LECTURES
ON
MORAL SCIENCE.
[UNEDITED]
DELIVERED BEFORE THE
LOWELL INSTITUTE, BOSTON.
BY
MARK HIOPKINS, D.D.LL.D.
PRESIDENT OF WILLIAMS COLLEGE; AUTHOR OF "LECTURES
ON THE EVIDENCES OF CHRISTIANITY," ETC.
BO S TON:
G O U L D A N D L I N C O L N,
59 WAS IN GTON STREET.
NEW YORK: SHELDON AND COMPANY.
CINCINNATI: GEORGE S. BLANCHARD.
1 8 6 3.
Entered according to Act of Congress, in the year 1862, by
GOULD AND LINCOLN,
In the Clerk's Office of the District Court for the District of Massachusetts.
SINCE 1830.
Permit me, my friends, a word of explanation with those of you who may read the following Lectures. It seems called for by the difference between them now, and when they were heard by the most of you.
In 1830 I was elected to the Professorship of Rhetoric and Moral Philosophy in this college, and during the first year prepared and delivered twelve Lectures on Moral Philosophy. Of these, omitting the introductory one, the first paragraph was the following: "If the human constitution
was made by a wise and good being, it must have been made for certain ends; and in those ends, whatever they may be, and nowhere else, can its perfection and happiness be found. To discover these ends and the means of attaining them, is the object of Moral Philosophy." Then followed such an examination of the constitution of man as I was able to make. This shows that the present lectures are but the carrying out of my original thought; but that those lectures should have been delivered for more than twenty-five years without essential alteration is what requires explanation, if not apology.
The explanation is, chiefly, from the pressure of other duties. During the remaining years of my professorship, my leisure was occupied with lectures on Rhetoric and Natural Theology, in connection with extra duties imposed by the declining health of Dr. Griffin. Subsequently, and till 1855.
AUTHOR'S ADDRESS.
those of you then here will remember our studies together in Anatomy, and Mental Philosophy, and Moral Philosophy, and Natural Theology, and Butler's Analogy, and Vincent. Add to these, preaching; the administrative labor incident to my position; the publication of between forty and fifty pamphlets, and of a volume on the Evidences of Christianity, and it may not seem strange that when the years came round, as they seemed to, with increasing rapidity, I was only able to give the lectures as they were.
Always feeling that my first duty was in the class-room, my strength simply sufficed for the demands of the passing day. In 1855 the Rhetoric of the class passed into other hands, but so much of work still remained that a revision of the Lectures was not undertaken till 1858. In the winter of 1861, the course, with the exception of the last lecture, for which there was not time, was delivered before the Lowell Institute.
When the Lectures were first written, the text-book here, and generally
in our colleges, was Paley. Not agreeing with him, and failing to carry out
fully the doctrine of ends, I adopted that of an ultimate right, as taught by
Kant and Coleridge, making that the end. If, therefore, any of you still
hold that view,-as doubtless many do,-it is not for me to say that you
have not good authority for it, or to complain if you object to that now
taken.
But whatever may be said of this central point, the Lectures have been
much changed in other respects, and, as I hope, improved. Such as they
are, with thankfulness that I am permitted to address so many of you, and
with many pleasant recollections of our former discussions on this subject,
they are now committed to your candid and indulgent consideration.
Your sincere Friend,
MARK HOPKINS.
WILLIAMS COLLEGE, OCTOBER 1, 1862.
P R E FACE.
PHILOSOPHY investigates causes, unities, and ends.
Of these it is the last two that are chiefly considered in
the following lectures. " Happy," it has been said, "is
he who knows the causes of things." But in a world
where there are so many apparent discrepancies both
natural and moral, he must be more happy who knows
the arrangement of things into systems, and sees how
all these systems go to make up one greater system and
to promote a common end. An investigation of causes
respects the past; of unities and ends, the present and
the future. Of these the latter are more intimate to
us, and he who can trace the principle of unity by which
nature is harmonized with herself, and man with nature,
and man with himself, and the individual with society,
IX
PREFACE.
and man with God, -who can see in all these a com
plex unity and can apprehend their end, -will have an
element of satisfaction far greater than he who should
know the causes of all things without being able to
unravel their perplexities.
From the place assigned to Moral Philosophy in the
classification adopted in these lectures, an incidental
consideration of the above harmonies seemed to be
required. Hence it is hoped that the book may con
tain suggestions that will be valuable to some who may
not agree with its doctrines on the particular subject
of morals. It is particularly hoped that it may do
something towards introducing more of unity into the
courses of study, or some of them, in our higher semi
naries. If the works of God, regarded as an expression
of his thought, are built up after a certain method, it
deserves to be considered whether that thought will not
be best reached by following in their study the order
that has been followed in their construction, and which
is involved in that method. Something of this I have
X
PREFACE.
long aimed to do in my instructions, and with very
perceptible advantage. With suitable text-books and a
right arrangement of studies, much more might doubt
less be done.
In treating of any natural system, as each part
implies all the others, wherever we begin, and what
ever method we follow, we are compelled to use terms
whose full meaning can be reached only in the progress
of the investigation. This is particularly true when, as
in the present instance, instead of beginning with defi
nitions, we seek for them. For this it is hoped that
due allowance may be made.
It will be seen that important, and even cardinal
points, are often but briefly touched in these discus
sions. I can only say that the work is, of necessity,
suggestive rather than exhaustive, and that if these
points are so treated as to show their place in the sys
tem, the outline may be readily filled up.
For remarks upon the present condition of the sci
ence, and for the general course of thought pursued,
XI
PREFACE.
the reader is referred to the opening lecture, and to the
summary at the close.
English literature is rich in ethical speculation. Sev
eral valuable treatises have recently been published in
this country; but the ground of classification, and the
general aspects an(d connections of the subject, as pre
sented in the following lectures, are so far different
from others, that it is hoped something may be gained
to the science by their publication. To the authors of
the treatises above referred to, and also to the friends
who have aided me by their suggestions, I desire to
express my indebtedness.
I will only add, that the work is written in the interest
of truth, and not controversially.
XII
C O N T E N T S.
LECTURE I.
MORAL SCIENCE AND ASTRONOMY.- REASONS FOR THE SLOWER PROGRESS OF
THE FORMER. — PROGRESS MUST BE SLOW. - TWO CLASSES OF SCIENCES.
-USE OF STUDYING THE SCIENCE,......................... o.. 17
LECTURE II.
THREE QUESTIONS. -THE CONSIDERATION OF ENDS.-AN END ATTAINED
IN THREE WAYS. -ENDS SUBORDINATE, ULTIMATE, AND SUPREME. -AN
END INVOLVES A GOOD. -THE NATURE OF GOOD AS FROM ACTIVITY. -
TIIE GREATEST GOOD.....................................39
LECTURE III.
KINDS OF GOOD. - SUSCEPTIBILITIES- AND POV.ERS. - GOOD AS HIGIIER
AND LOWER. - FORCES AND FACULTIES - THEIR SUBORDINATION. - THE
LAW OF LIMITATION.- METHODS OF ADDITION AND OF DEVELOPMENT. NATURAL AND CHRISTIAN LAW OF SELF-DENIAL............. 59
2 XIII
CONTENTS.
LECTURE IV.
RELATION OF INTELLECTUAL AND MORAL PHILOSOPHY. - SPONTANEOUS
AND VOLUNTARY ACTIVITY. - FACULTIES INSTRUMENTAL AND ULTI MATE. - INSTINCT.- THE APPETITES. - NATURAL - ARTIFICIAL. - THE
DESIRES. - CLASSIFICATION OF THEM. - DESIRE OF CONTINUED EXIST ENCE,........,................................. 79
LECTURE V.
DESIRE OF PROPERTY.- AVARICE.- DESIRE OF KNOWLEDGE.- DESIRE OF
POWER. - INFLUENCE. - EMULATION. - DESIRE OF ESTEEM. - DESIRE OF
GLORY...............................................102
LECTURE VI.
THE AFFECTIONS. - NATURAL AND MORAL. - BENEVOLENT. - DEFENSIVE
AND PUNITIVE. - ORIGIN OF MALEVOLENT AFFECTIONS.- FORGIVENESS.
- HOW SUBJECT TO WILL. - THE INTELLECT.- LOVE OF TRUTH,.. 129
LECTURE VII.
THE MORAL NATURE. - REASON. - IDEAS OF DIFFERENT ORDERS. - HAVE
AN ORDER OF DEVELOPlIENT. - FREE-WILL. - PERSONALITY. - ACTION
TO WHICH RE;SPONSIBILITY ATTACHES. -ALL MORAL PHENOMENA IN
xiv
I
CONTENTS.
CONNECTION WITH TIlE CHOICE OF A SUPREME END.-CONSCIENCE.THE MORAL NATURE DOUBLE.-TIHE HIGHEST GOOD.-COINCIDENCE OF
NATURAL AND REVEALED LAW........................ 157
LECTURE VIII.
RELATION OF VIRTUE TO HAPPINESS. -QUANTITY AND QUALITY OF GOOD
-MORAL AND NATURAL GOOD. - REGARD FOR OUR OWN GOOD. - CON NECTION WITH BENEVOLENCE. - ENJOYMENT FROM APPROBATION.
THE TRUE END OF MAN.-'CONNECTION BETWEEN MORAL AND NATURAL
GOOD,............................................181
LECTURE IX.'
THE SPHERE OF MORAL SCIENCE. - RIGHT AND WRONG. - DEFINITION OF
TERMS.- PROVINCE OF CONSCIENCE. - HOW FAR INFALLIBLE. -TWO
SPHERES. -DIVERSITY OF MORAL JUDGMENTS.- CRISES OF LIFE. - RE LATION OF CONSCIENCE TO OTHER PRINCIPLES OF ACTION. - COMPLEX ITY OF MOTIVES.- MUST A VIRTUOUS ACT BE FROM A SENSE OF
DUTY?............................................205
LECTURE X.
RECTITUDE AND VIRTUE.- RELATIONS. - EXPEDIENCY, PRUDENCE, AND
VIRTUE. -ORIGIN OF MORAL DISTINCTIONS AS RELATED TO THE DIVINE
NATURE. - COINCIDENCE OF INSTINCT AND REASON- OF FAITHII AND
REASON-OF PHIILOSOPHY AND) IELIGION.............. 228
X~
-CONTENTS.
LECTURE XI.
RIGHTS. - THEIR ORIGIN AND KINDS. - ALIENABLE - INALIENABLE. SLAVERY. - RIGHTS OF PERSONS AND OF THINGS. - GIVING AND RE CEIVING. -RIGHTS OF GOVERNMENT. -LIBERTY AS RELATED TO RIGHTS.
- DIFFERENT KINDS OF LIBERTY - NATURAL, CIVIL, POLITICAL,..253
LECTURE XII.
A FUTURE LIFE. - ITS RELATION TO MORALITY. - THE PHYSICAL ARGU MENT. - MORAL ARGUMENTS............................... 277
SUMMARY................................ 296
XvI
LECTURES ON MORAL SCIENCE.
LECTURE I.
MORAL SCIENCE AND ASTRONOMY.- REASONS FOR THE SLOWER PROGRESS OF
THE FORMER. - PROGRESS MUST BE SLOW. - TWO CLASSES OF SCIENCES.
-USE OF STUDYING TIHE SCIENCE.
AMONG the sciences which earliest drew the attention
of man were those of Astronomy and Morals. Of these,
one respects the sources of that light which is from without, the other of that which is within. Of the one, the
objects and phenomena are not only without us, but are
separated from us by inconceivable distances; of the other,
the phenomena are not only within us, but belong to that
part of our nature which is special to us, and whose circle
lies nearest to its central point.
Connected with each are practical judgments common
to all. Both the heavens and the moral nature of man
yielded him guidance before there was a thought of the
science of either. The unscientific man rejoices in the
light that comes from Arcturus no less than if he could
analyze its beams, and is guided by the polar star no less
surely than if he could measure its magnitude and distance. The day and the night, the changing moons and
2* 17
LECTURES ON MORAL SCIENCE.
the revolving seasons, are alike to all. In the same way
men agree in their practical judgments on the great subjects of morals. By their original nature there is within
them a guiding light by which the learned and unlearned
alike may walk. But in either case, when science began
its work, and asked for causes, and reasons, and classifications, there were conjectures and diversities of opinion
without end. Of the apparent movement of the heavens,
and of a virtuous or heroic act, men judged alike; of the
cause of that movement, or of the nature of virtue itself,
they did not judge alike. Practically, men could agree in
both; but in everything pertaining to the science of either,
nothing could be more discrepant than their opinions, or,
for ages, more discouraging and apparently hopeless than
the attempt to establish any one doctrine that should be
generally accepted.
If, now, the inquiry had been made in the early period
of these sciences which of them would soonest reach
perfection, the unhesitating answer would doubtless have
been,- that of which the phenomena are within us, which
are immediately testified to by our consciousness, and are
always subject to our notice. Whether man would ever
be able to perfect a science of the heavens, might well have
been doubted; but that he should do this sooner than perfect the science of that which pertained to his own most
intimate being, and which stood in the closest relation to
his highest interests, could not have been believed. But
so it has been. After ages of observation and conjecture,
during which the phenomena seemed in hopeless confusion; after exhausting the efforts of some of the best minds
in every age, the central truth of astronomy at length
dawned, and the chaos of conjecture became the order of
science. From a science of observation, astronomy has
18
MORAL SCIENCE AND ASTRONOMY.
now become one of deduction, and if not altogether complete, is more nearly so than any other.
The success thus achieved in the field of astronomy was
a great stimulus and encouragement to effort in other departments. From the vastness of its distances, the magnitude of its objects, the complexity of its phenomena, and
from its inconceivable forces and velocities, there was connected with success here an excitement and sublimity
which greatly heightened the purely scientific pleasure,
and which inspired a confidence of future triumphs in
whatever should be attempted. Nor was this confidence
without a basis. In the advance of every form of physical
science then known, no period of the world can be compared with that since the time of Kepler and of Newton.
Meantime, forms of science then unknown, as chemistry
and geology, have sprung into giant proportions; while
the application of science to the arts, employing every
substance, and harnessing every force in nature for the
service of man, is revolutionizing not only society, but the
face of nature herself.
In mental and moral science there has, too, been greater
activity than ever before; but we are not, perhaps, in a
position as yet to say how much there has been of progress. There are still discordant voices, and different
schools, and those that say "Lo, here," and "Lo, there;"
and perhaps the variety of systems proposed, especially in
morals, was never greater.
Thus situated, it is an encouragement to think of the
seas of doubt through which astronomy has waded. We
remember that the perplexities of its votaries were once
as great as ours can be now, and hope for a similar deliverance. The end of investigation is attained when we either
comprehend all that is brought before us, or can draw the
19
LECTURES ON MORAL SCIENCE.
line which shall fix the natural limits set by God to our
knowledge; and we are not of that desponding, or rather
indolent class, who distrust the powers of the human mind
to do, in all cases, one or the other of these.
So far as our present subject is concerned, it may aid us
in doing this if we inquire for a little how it has happened
that physical science, and especially astronomy, has so far
outstripped moral science. What are the causes of a result so impossible to have been anticipated?
And first, we may mention a difficulty much insisted on
by Chalmers, as pertaining to the observation of all mental phenomena. This arises from the fact that the mind is
both the observer and the thing observed, and that some
of its states at least (they say all) are of such a character as to preclude examination at the moment they exist.
Thus, when a man is thoroughly angry his whole thought
is directed to the object of his anger, and nothing can be
conceived more incompatible with the state of an angry
man than that he should be engaged in taking psychological observations on himself. The moment he turns his
attention from the object of his anger to himself for the
purpose of observing it, the anger is gone. It cannot,
therefore, be studied directly, as we study the objects of
our senses, but only as it is remembered.
This holds in all cases of violent emotion and should
have its just weight, but not in the ordinary states of
thought and feeling. If the view of Chalmers and of
Brown before him were adopted in its strictness, no man
would ever know his own present state, but only the
states he had been in, and so could never deal with his
present, but only with his past self. The moment his
attention should be so far called to himself as to inquire
whether he was angry, his anger must cease; and the
20
REASONS FOR ITS SLOWER PROGRESS.
prophet of old who thought he was angry, and said he did
well to be, was mistaken. In thinking, we know not only
the object of our thought, but ourselves as thinking. The
consciousness is so far complex as to embrace both. So in
the feelings. There is no more difficulty in supposing such
a complexity of the consciousness as to embrace both an
act and a feeling caused by an act, than there is in supposing that the same consciousness can embrace the remei -
brance of an act and the feeling caused by that remembrance.
There is doubtless at this point a real difficulty, but we
think it less formidable than it is made by Chalmers and
others.
To a successful investigation the first requisite is a clear
apprehension of the subject to be investigated as distinguished from everything with which it may be confounded,
or to which it is related. This discrimination in regard to
morals has often failed to be made. This is the second
reason.
Language accommodates itself, after a time, to the exigencies of thought; and when clear discriminations are generally or persistently made, there will be terms to express
them. In the Latin language, the word for conscience and
for consciousness was the same; it is so still in the French,
Italian, and Spanish, and this was formerly true of the
English. But if the moral consciousness were not now
partitioned off, and its phenomena grouped by a word of
its own, we may easily see how difficult it would be to disentangle those phenomena from the mass of other things
covered by the same word; and while the language remained in that state it was scarcely possible that much
progress should be made in the science. But as thought
was concentrated and analysis progressed, that which was
21
LECTURES ON MORAL SCIENCE.
consciousness par eminence, the moral consciousness, appropriated the term conscience; and yet no one can now
read even the scientific writers on the subject and not perceive that they still use the term with a wide diversity of
signification.
It was this state of the language, or more properly of
the public mind represented by it, that rendered possible
in the Scotch universities such a state of things as is come
plained of by Chalmers. Hie says: "In the hands of
some of our most celebrated professors, it" (i. e. moral
philosophy) "has been made to -usurp the whole domain
of humanity, insomuch that every emotion which the
heart can feel, and every deed which the hand can perform, have, in every one aspect, whether relating to moral
character or not, come under the cognizance of moral philosophy." He calls the science as there treated "a strange
concretion," "a vast and varied miscellany," which he
wished "to marshal aright into proper and distinct groups."
How this subject has been regarded in England we
may learn from an introductory lecture to a course on
Moral Philosophy delivered in London by Sidney Smith.
"Moral philosophy," he says, "properly speaking, is contrasted to natural philosophy; comprehending everything
spiritual, as that comprehends everything corporeal, and
constituting the most difficult and the most sublime of
those two divisions under which all human knowledge
must be arranged." "In this sense," he proceeds, "Moral
Philosophy is used by Berkley, by Hartley, by Hutcheson,
by Adam Smith, by Howe, by Reid, and by Stewart.
In this sense it is taught in the Scotch universities,
where alone it is taught in this island; and in this sense
it comprehends all the intellectual, active and moral faculties of man; the laws by which they are governed; the
22
REASONS FOR ITS SLOWER PROGRESS.
limits by which they are controlled, and the means by
which they may be improved." In accordance with this,
we find in his course, lectures on external perception, on
taste, and on wit and humor, while in his whole twentyseven lectures he did not treat of the conscience, or of
right and wrong, at all.
Such a blending of departments, all covered by one
name, in a single professorship, could not be favorable to
accurate analysis. There were reasons for it. Mental and
moral science are nearly related; but all knowledge is
related to all other knowledge at some points, and it would
be scarcely more incongruous to assign geography to the
astronomer because the earth is one of the planets, than
to group external perception and the knowledge of duty
under the same science because they both belong to the
mind.
A third cause of the slower progress of moral science
is its greater complexity.
All science supposes uniformity in the phenomena, and
so, in their cause or law, which is what science seeks. If
there be no cause acting uniformly, and tending to entire
uniformity of results, there is no basis for science. But
with such a cause, the complexity will be in proportion to
the number of disturbing forces that may come in between
it and the phenomena as seen by us. In astronomy these
disturbing causes are comparatively few. Gravitation
towards the sun only, would cause the planets to move
in a perfect ellipse. But none of them do thus move, and
it is obvious that disturbing forces might be multiplied so
as to render a science of the stars, or at least any other
than a hypothetical one, impossible. Here lies the obstacle
to a science of the winds. There is doubtless uniformity
of causation, but the phenomena, as known to us, are so
23
LECTURES ON MORAL SCIENCE.
modified that we cannot trace each one back to its cause,
or predict the future. So of human conduct. Men are
themselves unlike, and in endless variety. Motives are
complex. The effects of education, of social position, of
political institutions and of climate, are to be estimated;
and even though all the actions of men might be referred
to one principle, it would be impossible to trace them to
it, or to predict with certainty the course of any one individual under its guidance.
When we look, then, at this greater complexity and
remember that the study of processes within us, mental
and moral, is connected with no such pleasure as observation by the senses, and can have no such aid from others,
we find a reason of no little weight for the slower progress
of this science.
A fourth reason is to be found in the fact, which we
should not have anticipated, that the nearer we come to
that in our being which is most intimate and central,
which is our very self, the more difficult observation
and analysis become.
As early, certainly, as the time of Cicero, the mind was
compared to the eye, because that sees other things but
not itself. The power of making itself an object to itself
belongs to the mind of man as he is distinguished from
the brutes; it is the last power that is developed, and in
most men is scarcely developed at all. But where this
power is developed it begins with those phenomena which
are most outward and least essential. Hence, not only in
matter, but in mind, completed science will probably travel
from that which is more remote, or more outward, to that
which is nearer, or more inward.
It is now generally conceded that there are two kinds
of knowledge, or cognitions, - one which we gain of,
24
REASONS FOR ITS SLOWER PROGRESS.
and from, the external world through the senses; and the
other that which springs from the mind itself after its
powers have been waked into consciousness. It is not
supposed that there are innate ideas, but that the mind
has fixed capacities by which, in connection with the exercise of consciousness, it necessarily and universally forms
certain ideas and affirms certain truths. These ideas and
truths, if such there are, must be more intimate to us than
any other part of our mental furniture; but it is precisely
respecting these, and the field which they claim, that the
most subtle and difficult of all the problems in philosophy
have arisen. That we have ideas through the senses no
one has ever doubted, and they are readily classified and
their characteristics given; but nothing could more strikingly illustrate my present point than the fact that the
very existence of any such truths and ideas as those just
mentioned has been doubted, and still is. The reception
or rejection of these cognitions as elements of philosophy
has been the dividing line between its different schools
from Plato down. Probably the preponderance in numbers has been against them, and even now they are rejected by such men as Comte and Mill.
As we should anticipate firom the fact just stated, the
advocates of these cognitions have failed to give their
characteristics, and thus to bring them out into distinct
consciousness. Before Leibnitz, no one had ever mentioned their two great characteristics, - necessity and Universality, -and it was not till the time of Kant that
these were at all signalized and properly applied. Meantime, there was no uniform and accepted designation
either for the cognitions, or the faculty in which they originated. The faculty was called "intuition," and the "dry
light of the mind," and "common sense," and "the
3
25
LECTURES ON MORAL SCIENCE.
reason," and by Hamilton it is called "the regulative faculty," while the names for the cognitions themselves were
still more numerous.
But remarkable as all this is, it is still more so that no
one has even claimed to explore all the recesses and sound
the depths of this faculty. Some ideas, as those of existence, identity, and space, are recognized by all of this
general school as given by it, but no one has claimed to
make an accurate and full statement of these native, necessary, and universal cognitions. They have lain, and still
lie, like a nebula in the depths of the heavens, which no
instrument has as yet been able fully to resolve.
Among and concerning these it has been that the great
battle with skepticism - that is, philosophical skepticismhas been fought. Hume denied their validity. Their
legitimacy and place was not recognized in formal logic,
then the test of truth, and the mass of philosophers were
in the unfortunate position of holding to principles clearly
involving consequences which they could not accept.
Skepticism had thus an apparent triumph. Meantime,
Reid began groping about in this region, and found the
means, as he and others thought, of bridging the chasm of
inconsistency dug by the skeptics; but so great was the
want of precise terms, and so subtle the elements he dealt
with, that even the acute Brown not only did not comprehend him, but imputed to him opinions the very reverse
of those he held. In such a state of terms and ideas, men
are like Indians fighting in a thicket. It is not easy to
find and dislodge your adversary; and when you do, he
can easily gain another place of concealment, and deny
that he has been dislodged at all. If a clearexposition of
these truths of reason, or native cognitions, or first truths,
or maxims of common sense, or fundamental laws of be
26
REASONS FOR ITS SLOWER PROGRESS.
lief, or whatever we may choose to call them, could have
been made before the time of Hume, he would probably
never have been heard of as a philosophical skeptic. The
mind of Hume had in perfection the acuteness of the
skeptic, which enabled him to see defects, and so to
destroy, but had not the comprehensiveness needed for
construction.
But to take a plainer case. What can be more intimate
to a man, or more perfectly known, than that of which he
is conscious? If a man cannot know what he is conscious
of, it would seem that he cannot know anything; and yet
the whole question, between Reid and Hamilton on the one
side, and the great mass of philosophers on the other, respects simply the fact whether there is or is not given in an
act of consciousness, both a subject and an object that are
not, in the last analysis, identical. What consciousness testifies to must be accepted. This all allow. Not to do it
would be suicidal even to the skeptic; for he would have
no ground for affirming that he doubted. The only ques- ~
tion is, what it is that consciousness gives. If we say that
it does thus give both the subject and the object, that sim
ple affirmation sweeps away in a moment the whole basis
of the ideal and skeptical philosophy. It becomes as the
spear of Ithuriel, and its simple touch will change what
seemed whole continents of solid speculation into mere
banks of German fog. If we say that the subject and
object are not both given, we are then left to find as we
may a solid basis for our belief in the existence of an
external world. But however we may decide it, the fact
that thie great philosophical dispute of the day would be
settled at once by a precise statement of what is given in
the consciousness of every man, shows clearly that our
investigations become more difficult as we approach the
27
LECTURES ON MORAL SCIENCE.
centre of our being. It shows, too, how far apart, on
subjects like these, men may be in their statements, whose
belief is really the same; for the consciousness is really the
same in all, and is accepted by all.
What has now been said relates indeed to the intellect;
but the moral nature is not less central, and presents, to
say the least, equal difficulties on this ground.
In connection with the above, it may be well to notice
a peculiarity of all advances and discoveries made in this
direction, as they are related to the mass of men who are
not philosophers. It is, that the more profound and difficult the discoveries are, the more they will seem, when
clearly announced, to be a matter of course, and no discoveries at all. Though few men are able to state what is
really contained and implied in their own consciousness,
yet, when it is stated by another, there can be in it nothing
strange to them; they recognize it and say, "Yes," "Certainly," and it seems to them they could have made the
same statement. The continent is discovered, the egg is
set on end, and nothing could have been easier.
A fourth reason, which has been implied already, and
which has operated both as cause and effect, has been
the want of definite terms. Science requires that terms
should be used uniformly in the same sense, and that they
should convey the same impression to all who use them.
This can be done perfectly only in mathematics, may be
approximated in dealing with objects of sense, but is most
difficult in all that pertains to mental and moral science.
In these the terms are borrowed from material objects, and
so can be applied only figuratively; and then in applying
them there is a difficulty that does not belong to physics.
When I point a child to a particular star and say, "That is
Jupiter," I am sure that we both see the same object; and
28
REASONS FOR ITS SLOWER PROGRESS.
when speaking of it thereafter, we cannot fail of understanding each other. But when I speak to him of "the
reason," as distinguished from "the understanding," or of
"first truths of reason," as distinguished from "empirical
knowledge," or of conscience, I speak to him of what is in
my own mind, and he must respond respecting what is in
his mind. But differing as we do in age, constitution, and
education, we can never be sure that our impressions are
alike. "What," said a master to his mail who had refused
to do his bidding on the ground of conscience, -" what do
you mean by conscience?" "It is," said he, "something
in here that says, I won't." In the opinion of Paley, if
conscience be anything original and native to the mind, it
cannot be distinguished from prejudices and habits. Some
think it simply the power of moral discrimination; others
add an emotive element to reward and punish, and others
still an impulsive power. Some regard it as the voice of
God, and nearly or quite infallible; others as simply a form
of judgment, like any other, and equally liable to error.
Hlere the same word is used; and so it is, only with a difference of meaning somewhat wider, when we speak of a
sign of a tavern, and of a sign of the zodiac; and till there
is agreement in the meaning of the term, no progress can
be made in discussion.
Hiow, then, shall we be freed from this difficulty? Who
shall have the right to say what the term conscience shall
include? No one. But as thought is concentrated, as it
will be, more and more upon man himself, the facts of his
moral nature will come into more distinct consciousness,
and the discriminations thus made will demand the narrowing down of old terms, or the invention of new ones,
and these will gradually become definite and generally accepted. When terms are thus gained, they will react upon
29
3*
LECTURES ON MORAL SCIENCE.
thought, as instruments invented react upon the inventive
power; for language is not only a product, but an instrument of thought.
This process is going on, slowly but satisfactorily, in
moral, and particularly in mental science. In the latter,
the old classification of the faculties was into those of the
understanding and the will. This sufficed till further examination showed that all the facts could not be ranged
under these. We not only think and will, we also feel;
and accordingly, after long discussion and some aid from
abroad, the threefold division of the faculties into those
of Thought, of Feeling, and of Action, is almost universally
accepted. Under each of these a distinct science, or, if
you please, department, is formed, in which a similar process must go on. In that of thought, or, as he terms it,
of cognition, Sir William Hamilton has introduced new
terms and classifications, some of which will doubtless be
adopted. The same will be done in the other departments, till the whole shall assume all the definiteness and
completeness of which the science, from its nature, is
capable.
This difficulty from a want of terms, and of uniformity
in their usage, has been felt from the first, and will be
appreciated the more as the subject is more studied. It is
one concerning which it is easy to give precepts, but difficult to follow them. Of this difficulty no one has been
more fully aware than Locke. iHe wrote largely upon it,
and gave wise precepts; and yet used the word idea so
loosely that on the great subject of the origin of knowledge it is still uncertain what he really believed. On the
Continent he was so interpreted as to be made the father
of materialism. Many of the English admitted of no such
interpretation, and both parties sustained themselves by
30
REASONS FOR ITS SLOWER PROGRESS.
adequate quotations. On this subject I make no promises.
The next, and the last consideration I shall adduce to
account for the slower progress of moral science, is the
failure of men in practical virtue.
That there has.been such failure no one will deny; and
it has operated in various ways. When a science, as formerly that of war, is regarded as at the basis of the great
business of life, it will be studied. Attention will be concentrated upon it, and it will be carried to the greatest
possible perfection. But let the subject be one for which,
while every one acknowledges its importance, few have
any practical regard; let it be thought of as something
which will do for the closet and the schools, but not for
practical life; let there be a general impression that its
maxims are repeated in a perfunctory way, as a cover to
the real principles of action, and any earnest or general
study of the science is impossible. Theories there may be.
They are needed for conversation and the Reviews; but
only as there is a real, heartfelt, practical interest in virtue,
can there be a genuine struggle for the truth as vital. The
general failure of men in practical virtue has created an
atmosphere unfavorable to an earnest search after the
truth in morals. The set of the current in society has
been against it.
Again, under this head, in proportion as men are vicious,
not only will they lack interest in the science, but they
become disqualified for its pursuit.
This is in accordance with the laws of all the faculties.
The faculties are strengthened by exercise; they can be
strengthened in no other way; and they are exercised rightly only by doing just the work which God intended they
should do. The moral powers, as a whole, can be so exer
31
LECTURES ON MORAL SCIENCE.
cised as to improve them only as duty is accepteed and practically performed. Therefore we say that the man whose
moral faculties have been dwarfed by disuse, or perverted
by abuse, would not be well qualified to investigate the
science of morals. The phenomena, it is to be remembered, are those of mind. While there is, therefore, in all
a common basis for the science, yet both the seeing eye
and the thing seen may be modified by custom and habits.
If there be little moral culture, the moral phenomena will
either be obscure, or will consist in a decided wickedness
which is blinding and hardening, while, at the same time,
the power of moral discrimination will be diminished. It
may be said that it is the intellect that constructs science.
But it must construct it out of the materials given, which
will be different in a vicious mind; and it must also have
clearness and power in the particular field in which it
works. But no fact can be better established than that
wickedness, in every form and degree, not only blunts the
moral feelings, but weakens the power of moral discrimination. A perfect mental science would require, first, the
normal action of the faculties to give the phenomena, and
then an accurate observation of those phenomena. A perfect moral science would require the normal action of the
moral powers, either in ourselves or another, and an accurate observation of the results; but by the prevalence of
vice the facts are both distorted and obscured.
What has now been said of morals is equally applicable
to taste. A man whose sense of beauty should be either
uncultivated or perverted would be the less capable of
apprehending and presenting perfectly the science of esthetics. But there is in morals a special difficulty. A
vicious man is strongly tempted either to deceive or to
bribe his conscience, and can hardly be expected to judge
32
PROGRESS MUST BE SLOW.
fairly of any system that would either justify or condemn
himself. In all moral and religious truth there is this difficulty. It is not that we have lost the power to judge, but
that we will not use it. It is the old difficulty of the influence of the desires and affections and of our supposed
interest on our judgment. We all know how passion and
interest pervert the judgment, and what discordant opinions there are wherever men are under their influence.
If; therefore, there had been no incapacity from vice,
and no wrong bias, or passion, or want of candor, we cannot but suppose that moral science would have been much
more advanced than it now is.
In thus answering the natural inquiry respecting the
relative progress of these sciences, I have desired also to
do something in the way of caution and guidance for ourselves. What has been is now, and will continue to be.
The same obstacles that have been encountered by others
we shall encounter; and sown of them are such that if we
are forewarned we may be forearmed against them.
Against the first difficulty mentioned we can do nothing
directly; but it is a satisfaction, and may be of some aid,
to know the precise'mode in which our observations are
to be made. But we may gain definite views of the
sphere and objects of the science; we may seek to simplify it; we may make independent search into the depths
of our own consciousness; we may be careful in the use
of words, conforming at least to our own definitions; and,
above all, we may either enter upon, or become more earnest in, a course of practical virtue, and so both prevent
the imbecility of vice, and disperse the blinding mists that
always arise from a corrupt heart.
The difficulties just considered are such as to preclude
the hope of any great and sudden advance in this science
33
LECTURES ON MORAL SCIENCE.
-of any, at least, which shall be at once recognized and
incorporated into the public mind. Even if completed in
thought and expression by one man,-if it should have
its Newton, -yet its full acceptance by the public mind
and assimilation with it would necessarily be slow. In
astronomy, the true system was opposed to the popular
conceptions and forms of speech, and more than one generation was required for it to permeate the masses and
thoroughly control the habits of thought. But in that
the proofs were open to popular apprehension, and, for
the most part, there were no desires and passions to obstruct conviction. But of all the changes in society, none
are so slow as those which are conditioned upon changes
in language and character. Even Christianity itself, with
its wonderful evidence and its divine power, is far from
having taken full possession of the public mind in any
community, and simply because it had these obstacles to
encounter. But, as we h-aveaseen, perfection in moral science, to say nothing of other obstacles, can be reached
only through changes both in language and in character.
If terms absolutely new would not be demanded, yet some,
like the heathen words for God, would require to be expanded and ennobled, while others would require to have
their elasticity and capacity reduced; and then, the delicacy of moral feeling and accuracy of perception to be
attained only through virtuous habits, would be indispensable.
It follows firom this that, in our cultivation of this field,
we are not to be disappointed if we see no immediate or
startling results. The clianges to be anticipated are like
those of geology ill the formation of strata, - sometimes
more and sometimes less rapid, but always relatively slow.
But since the progress of the science is so slow, and its
84
TWO CLASSES OF SCIENCES.
completion has been so long delayed it may be asked
whether it is of any use. Are there not, it may be inquired, in our nature practical principles, which do and
will control the course of human affairs with something
like the certainty of instinct, and quite independent of
scientific speculation? Within the memory of many this
question has been put respecting various branches of physical science. It would not be put now. But respecting
metaphysical and moral science there are those who put it
with sincerity and earnestness.
On this point, and as they are related to practical arts,
there are two classes of sciences. In the one the science
is wholly at the basis of the art, and is requisite to its
results in any degree. The art of photography could not
have been without chemistry, nor surgery without anatomy, nor the art of protecting buildings from lightning
without the science of electricity. In such cases, and they
are numerous, the science is first, and the practical results
follow. The processes start fiom the sciences. In these
cases no one questions the utility of science. In the other
class the practical results are first and the sciences follow.
The sciences start from the processes, which they simply
recapitulate. Here science consists in the statement of
the properties, the relations in space, and the successions
in time of those things which our will cannot reach, or, if
it can, cannot improve. Science may predict the place of
a star; but the color of its light or the rapidity of its
motion it cannot affect. God gives light and the eye, and
we see; but we see no better after knowing the structure
of the eye and the science of optics than before. Here
the result is first, as perfect as it can be made, and the
science- is just a statement of the process by which the
result was reached. It is in this class that the science of
35
LECTURES ON MORAL SCIENCE.
the mind belongs. Like the eye, its faculties are given,
and act by their own law without reference to science,
which can merely trace back and state such results as are
common to all minds. It is solely with reference to these
sciences that the question arises.
To this question, What is the use? there are two replies. The first is, that, even in the sense of the word as
used by the objector, these sciences are of use. The processes may be perfect; we may not be able to affect the
results, and yet the sciences may be of use indirectly.
We cannot change the number or movements of Jupiter's
satellites; but by means of their eclipses we can calculate
the longitude. Entomology will not enable us to change
the structure or habits of an insect; but it may suggest
a mode of saving our trees. The laws of the winds are
not subject to our control; but by a knowledge of them
we may shorten our voyage.
Again, this class of sciences becomes greatly useful
when the structures and processes of nature become deranged. When the eyes become flattened by age, science
can remedy the defect, and when
"A drop serene hath quenched their orb,
Or dim suffusion veiled,"
it is by science alone that it can be removed. And so it is
in most cases of displacement and derangement in the
physical system. The science of anatomy, which is almost
wholly at the basis of the art of surgery, would be of no
practical use if nothing ever went wrong in the body.
A second reply to the objection urged is, that while we
do not repudiate the conception of utility involved in
what has just been said, we yet do not need it for the
vindication of these sciences. We are capable of an inter
36
I
USE OF STUDYING THE SCIENCE.
est in science for its own sake, which shows that we have
an affinity with higher natures, and that the whole domain
of the universe will finally be ours. The pleasure felt by
the great discoverers of scientific truths is among the purest and most unselfish that can belong to man. It gives
them a thought of God which they utter to the race, and
it becomes a fountain of joy to millions. So it has been in
astronomy. Every time the thought of God, as uttered by
Newton, has been apprehended for the first time by any
mind, there has been a commencement anew of.the revolutions of the heavens, and the morning stars have sung
together. And so would it be if the mighty forces and
bright order that are without and afar could be seen to be,
as they are, but a type and reflection of the forces and
order within. Then would the great thought of God, not
merely of a physical order in one department, but in all
departments; and not of a physical order merely, but of
one which should correspond in his one universe to a
spiritual and moral order still more glorious, stand fully
revealed, and should be a light and a joy forever.
I have only to add, that our opinions of the laws and
processes of our being may be so held as to affect those
processes scarcely at all, and hence that the practical
results of man's opinions on these points are often less
beneficent and less mischievous than would naturally be
supposed. In our minds, no less than in external nature,
the forces are what they are, irrespective of any opinion of
ours, and will act, and no theory has any direct tendency
to eradicate or diminish them. In the man who believes
in disinterested benevolence, the desires and passions and
selfish forces may still have the ascendency, while he who
holds to the selfish theory may be often moved by the natural impulses of benevolence and sympathy. So it is that
4
. 37
LECTURES ON MORAL SCIENCE.
the selfish theory of morals, so long inculcated, has not
wholly corrupted society; so it is that men are often better
and worse than their theories; so it is that God holds
in check the evils that would naturally flow from the errors
of men.
88
LECTURE II.
THREE QUESTIONS.-THE CONSIDERATION OF ANDS.-AN END ATTAINED
IN THREE WAYS.-ENDS SUBORDINATE, ULTIMATE, AND SUPREME.-AN
END INVOLVES A GOOD.- THE NATURE OF GOOD AS FROM ACTIVITY. THE GREATEST GOOD.
THE difficulties mentioned in the last lecture as obstructing the progress of moral science would also render it less
desirable as a subject for a course of popular lectures.
But with those difficulties it has two decided advantages.
The first is, that it appeals directly to the consciousness
of the hearer. - No learning is needed; no science, no apparatus, no information from distant countries. "It is
nigh thee, even in thy heart." Some familiarity with
terms may be required; but there is that in every man
which may, and ought to make him a competent judge of
all questions pertaining to this science. Let a lecturer but
state the facts simply and truly, thus interpreting every
man's consciousness to himself, and he may hope to carry
his audience with him. Thus to state the facts will be my
endeavor.
The second advantage is, that the questions involved are
among the deepest and most vital that belong to our
nature.
We proceed, then, to our subject, and begin with facts.
That men regard some actions with approbation, and
others with disapprobation, is a fact, just as it is a fact that
they regard some portions of matter as hard, and others as
soft. Of those actions which they approve, they say that
39
LECTURES ON MORAL SCIENCE.
they ought to be done; of those which they disapprove,
they say that they ought not to be done.
In these facts we have our subject. Moral philosophy
respects the character and conduct of men only as there
are acts which they ought, or ought not to do. Wherever
the word ought, indicating duty, can go, there is its domain; and the point beyond which that cannot go fixes
its limit. Whoever can answer, in all cases, these three
questions, 1st. WVhat ought to be done? 2d. WhFy ought
it to be done? and, 3d. Iow ought it to be done? has
mastered the science of morals.
In answering these questions we may seek aid in various
directions. I propose to inquire, at present, what aid we
may derive from a consideration of ends as they stand
related to all rational and moral action.
In acting morally, man also acts rationally; but it is the
characteristic of rational action that it involves the conception of an end. Except in the apprehension of an end,
there is nothing that a rational being can do, or that a
moral being ought to do.
This relation of an end to all rational action may be
seen if we observe what occurs in the production, study,
and use of works produced by design.
In these the designer first conceives of the end, and then
of the thing designed with reference to that. It is, therefore, the end in view that controls the structure.
In studying a work produced by design, we may first
gain a conception of the structure, and pass from that to
the end; but our interest in the study of the structure is
from its apprehended relation to an end; and we are
never satisfied with a knowledge of structure without that
of the end.
The perfection of a work of design must consist in its
40
THE CONSIDERATION OF ENDS.
adaptation to attain its proposed end; and all use of it
except for this end must be either inappropriate or wrong.
Hlence, a conception of the end must control not only the
structure, but the use.
If the relation between the structure and the end be at
once perceived, there will be no need of rules. If not,
rules may be needed. These must grow out of the relation of the structure to its end, and will always express
some mode in which the structure must be used to attain
the end.
What is true of rules is true also of laws. These have
often been confounded, but are essentially different. A
law is imperative; a rule is not. A law has sanctions; a
rule has not. A law tells us what to do; a rule, how to do
it. A command to put forth continuous action directly, and
without the use of means, as to love God, would be a law,
but not a rule, and no rule could be given by which we
could do it. But though there are laws which cannot become rules, yet rules may become laws when the observance of them is commanded, and enforced by a penalty.
While, therefore, a rule prescribes a course of action that
would lead to an end, a law may prescribe one that is itself
an end. But even then, as a rule derives its value from its
relation to an end, so must a law derive its value from
what it is regarded as an end.
Again, regarding man as a moral being, if no end valuable in itself be supposed, it will be found impossible to
conceive of him as under obligation to act in any particular way. For the very conception of obligation that of an
end is a condition.
We see, therefore, that in all rational action the central
c6nceptio is that of an end. In works of design it controls the sructure in the mind of the designer; it is essen 4#
41
LECTURES ON MORAL SCIENCE.
tial to a right understanding of the structure by him who
would study it; it is in its relation to this that the structure
has its perfection and appropriate use; and from this that
the value of all rules and laws for, and in its use arises.
Of whatever can be comprehended and used, even of man
himself, all this may be affirmed.
Let us, then, apply these principles to man.
As man was made by a wise and good being, he must
have been made for some end, and the conception of this
end must have controlled the formation and adjustment of
every part of his complex structure.
From the study of this structure we may gain some
knowledge of its end. Aside from revelation, this is our
only knowledge on this point. Nor is the amount of
knowledge to be thus gained small. Of some parts of the
body, as the hands, the feet, the eyes, the teeth, the end is
revealed at once in the structure. It is this knowledge
of structure as related to use that gives comprehension.
Only in the light of it can we have complacency in the
structure when right, or the power to correct it when
wrong. In the same way the faculties of the mind, in
their relation to each other, reveal their end, and thus the
law of their use. An intelligent being whose end should
not be revealed in itself would be an absurdity. If the
end were not revealed to itself, it would be lost. It is the
possibility and measure of such knowledge that determines
the possibility and measure of any philosophy of man.
The perfection of man, viewed merely as a product of
divine power, must consist in his adaptation and capacity
to attain the end for which he was made.
That, and that only, is the right use of the faculties of
man, - of all his susceptibilities and powers of agency,by which they attain the end for which they mere made.
42
AN END ATTAINED IN THIREE WAYS.
If man could see the end for which he was made as God
sees it, and dispose himself perfectly for its attainment, he
would be in harmony with God; his faculties would work
in harmony with each other, and he would do all that he
ought to do.
Laws and rules for the conduct of man, whether confessedly human, or claiming to be divine, are valid only
as they are based on a true perception of the relation
between the human constitution and its proper end. If
a course of conduct, prescribed by what claims to be law,
could be shown to be opposed to the attainment of the
end for which man was made, it would not be right to
pursue it. The will of God is revealed in the end, and he
cannot contradict himself.
In the following discussions the word end, as applied to
man, will be of frequent use, and, to avoid ambiguity, it
may be well to say that the idea it involves is complex.
As intelligent and responsible, the end of man is to choose
something; as an agent, it is to do something; as capable
of enjoyment, it is to enjoy something; and as a creature
made by God, his end is to be that which will enable him
to do and to enjoy all that God intended he should. He
is to be something, to choose and do something, and to
enjoy something; and his whole end will be, to be what
God intended he should be, to choose and do what He
intended he should choose and do, and to enjoy what He
intended he should enjoy. He who should fail in any of
these would fail of attaining his whole end; and if the
word should at any time seem to refer particularly to one
of these elements, it will not be to the intentional exclusion of the others.
An end may be attained in three ways. And,1st. It may be attained by instinct. Here the agent
0
43
LECTURES ON MORAL SCIENCE.
has no knowledge or rational comprehension of the end,
but is guided by a blind impulse.
2d. An end may be attained by obeying a rule implicitly. Here there may, or there may not be a conception
of the specific end, but the connection between the means
and the end is never seen. In this way children are governed. Here the principle is not instinct, but faith. They
follow the rule, that is, they do as they would if they understood the connection between the means and the end,
and so receive the same benefit. To a finite being faith is
a necessary principle of action, and becomes practical wisdom when there is a rational ground of confidence in the
word of another because it is his word,-or of implicit obedience to his commands.
3d. An end may be attained understandingly and rationally. The structure may be known, now regarded simply
as a means; the end may be known; and there may also be
a clear perception of the mode in which the structure must
be used to attain the end. In this mode of action man
would not act from law, but from a knowledge of the
reasons of the law. He would be wholly a philosopher.
Viewing the end as God views it; voluntarily choosing
this end; applying all his powers as they were intended
for its attainment, he would do all that he ought to do,
would have the approbation of God, the approbation of
his own conscience, and the sanction of reason.
But if, in thus attaining the end for which he was made,
man would, as has just been said, do all that he ought to
do, then have we answered, in a general way, one of the
questions mentioned above. Would he thus do all that he
ought to do? If we say yes, then Moral Philosophy will
be the science which teaches man the end for which he was
made, why he should attain that end, and how to attain it.
44,
MORAL SCIENCE AND CHRISTIANITY.
To the above definition it may be objected, that it includes theology and religion. It does so only so far as to
justify a consideration of our duties towards God, and
that is found in all treatises on morals. Moral philosophy
differs from theology in taking for granted the being and
attributes of God; and religion differs from morality because it includes all our duties towards men as commanded
by God; and also because it implies an order of faculties,
and a class of duties, as those of worship, of which mere
morality could know nothing. Still, the science of duty,
of obligation, must be one. No satisfactory account of the
moral nature of man and of its full sphere can be given
on any other supposition. We may, if we choose, divide
our duties into those towards God, and those towards
man; but moral science must go wherever the word ought
can be applied.
But if not faulty on this ground, the definition has an
advantage in regard to Christianity. We are able, in the
light of it, to state precisely, which has not always been
done, the relation between Christianity, as a remedial
system, and moral philosophy. This is entirely different
from that of natural religion. It is that of medicine to
physiology. Physiology can know nothing of medicine
except as it would restore the system to health; and
moral philosophy, if we accept the above definition, can
recognize as obligatory no precept peculiar to Christianity, except as it can be shown to be necessary, in our
present state, to the attainment of the end for which man
was made. Let this be shown of any such precept, and its
obligation will not only be recognized, but it will be an evidence that the religion is from God; and a demonstrated
capacity in the religion to bring man fully to his end
would be a demonstration of its truth. From the consti
45
LECTURES ON MORAL SCIENCE.
tution of man moral philosophy would find his end. In
the end it would find revealed the will of God, and in the
relation between the constitution in its various parts and
its end would find revealed the law of God, and those
rules in accordance with which his faculties must act for
the attainment of the end. Christianity, on the other
hand, is wholly remedial. It supposes man to have broken
law, and it harmonizes with moral philosophy and can be
accepted by it only as it can attain for man his original
end, -or, it may be, something better, though of that
moral philosophy could know nothing.
Shall we, then, accept the above definitions? Let us do
so, so far, at least, as to make further explorations in this
direction. The definition speaks of an end; but ends are
of different kinds, and these we shall need to consider.
An end may be subordinate, ultimate, or supreme.
A subordinate end is one chosen for the sake of some
end beyond itself Thus books are chosen for the sake of
knowledge, and the implements of husbandry for the sake
of a crop. A subordinate end, regarded by itself, is not
necessarily-a good. It may be the reverse.
An ultimate end is one that is chosen for its own sake,
and without reference to anything beyond. It must be
some form of good. The enjoyment there is in viewing a
beautiful prospect is valuable for its own sake, and is the
ultimate end in making the ascent whence the prospect
may be seen.
An ultimate end, it is to be noticed, is always the result
of action, and never the action itself. It never lies proximate to the volition, and so cannot be the immediate
object in any act of volition, and is never commanded.
The formula Is —" do this and live." It is the thing to be
done that is commanded, and that is to be willed; the
46
VOLITION AND AN ULTIMATE END.
living is the result, and the ultimate end. So it is in everything. Eat the peach, -and enjoy it; take the remedy,and get well; ascend the mountain, turn your eyes upon
the prospect, - and enjoy it. The ascent and the turning
of the eyes are the immediate result of volition the enjoyment is not willed, but comes of its own accord as a result
of the constitution of the perceptive powers and the landscape in their relation to each other. It is here as in the
machinery which man constructs. He may build a mill,
supply it with wheat, set it in motion, and to all these
volition is in immediate relation. But the ultimate end
of the mill is the flour, and that is ground by the mill.
To that the will, as an executive act, is not proximate.
Hence, ultimate ends, those ultimate for man, have no
exchangeable value. They cannot be bought and sold,
and in this sense are worth nothing. As the brain has no
sensibility itself, but is the condition and fountain of sensibility for all other parts, so these, having no exchangeable
value, are the condition and ground of all such value.
Hence, after having chosen an ultimate end, an act, not of
volition, but of choice, we are always to understand what
it is that lies proximate tothat, and to attain that must be
the object of all immediate volitions and efforts. And
hence, again, in accordance with our present scheme of
thought, virtue will consist in the choice of the right end,
followed, of course, if the choice be thoroughgoing, by the
willing of that state or mode of activity which is believed
to be proximate to that. That state is always proximate
to the will; no means are required, and so a failure to be
virtuous admits of no excuse.
This relation of volition to an ultimate end has not generally been stated with sufficient distinctness, and the
result has been a constant puzzle. It is generally said that
47
LECTURES ON MORAL SCIENCE.
all men seek happiness, and yet no man ever made it the
direct object of volition. No man can. That God holds
in his own power. It is his immediate gift through that
constitution and relation of things which he has established. We will that which he has made the antecedent
and condition of happiness, and he gives the happiness.
We say "open sesame," and the gate opens of its own
accord. This is what men mean when they say they will
do their duty and leave the event with God.
But besides an ultimate, there is also a supreme or chief
end. A supreme end is also ultimate; but is that to which,
in any possible conflict of ultimate ends, all others must be
subordinated. Ultimate ends often, and necessarily, conflict with each other. The pleasure firom each sense is
ultimate; but it may be necessary to choose between that
of music through the ear and that of beauty through the
eye. In such cases we may indulge our preference; but no
end may be chosen as ultimate when it would conflict with
that which is supreme.
Any ultimate end may be adopted as supreme; but the
wisdom of man consists in choosing that intended by God,
which can be but one, and in giving to each of those
thus made secondary its proper place.
The choice of this supreme end is the highest act of a
rational being, and involves the activity of all his rational
and moral powers. It is the characteristic of a rational
action that it involves the conception of an end, and of a
moral action that it involves the preference of an end.
And as we regard a moral being as virtuous or vicious
according to' the end chosen, so do we regard a rational
being as wise or foolish on the same ground. Wisdom and
folly chiefly respect the ends which men pursue, rather than
the means by which they pursue them. Here, then, we
48
A GOOD.
find a point at which the rational and moral natures coa lesce.
But to be more particular. What the supreme end of
man, as fixed by God, must be, will be determined by what
he is in himself, and as related to other beings. The con ception and choice of such an end will therefore imply a
knowledge, implicit or formal, of himself, and of those
beings and relations through which alone the end can be
realized. This is the highest of all knowledge. There is
in it the rFOt aeav6p * of the ancients, and something
more.
In the conception of an end there is also involved that
of some good. This cannot come from the intellect alone.
There must be the activity of the emotive nature, - of that
through which we enjoy, as well as of that through which
we apprehend. But the recognition of a good through the
intellectual and emotive nature acting conjointly, does not
make it an end, much less a supreme end. To become
such it must be chosen. This involves the moral nature,
since the character of every man is determined by the end
he chooses. But, further, that a good should become a
supreme end implies that the will shall at once put forth
determinate acts for its attainment. Thus the conception
of a supreme end involves that of the action of the intellectual, the emotional, the moral, and the executive powers,
that is, of the whole personality,- of the man in his unity.
An end, as has been said, involves some form of good.
We next inquire, then, what is a good?
A good must be either some form of enjoyment, satisfaction, blessedness; or that which is the occasion, cause, or
ground of such enjoyment.
There are many objects without us so related to our
organs and faculties that enjoyment is the result of their
5 * Know thyself.
49
LECTURES ON I MORAL SCIENCE.
reciprocal action. Thus light acts upon the eye, and is
the condition of seeing. HIere we have the eye, the light,
and the product of their joint action, that is, seeing. A
peach eaten acts upon the palate, and is the condition
of a pleasant taste. IHere we have the palate, the peach,
and the result. Are, then, the light and the peach a good?
As conditional for these results, they are good, but not a
good. When we apply the term good, we mean either to
indicate that which is good in itself, or we have reference
to an end, so that the question may be asked, Good for
what? and if that question can be answered by indicating
any use of the thing for an end beyond itself, then it cannot be, so far forth, a good, nor can it be any part of a
supreme good. But all outward objects, and all possessions, sometimes called goods, have a use beyond themselves. If they were never to contribute to comfort,
enjoyment, or utility in any way, they would be goodfor nothing. It would seem self-evident that light never
seen, the sapid quality of the peach never tasted, would
not be a good.
We seem, then, driven, in our search for a good, to living, sensitive, conscious beings, whose faculties are in
action. If there were no consciousness in the universe,
there would be no good. But if found here, good must
be either in some state of the being that is back of the
activity; or in the activity itself; or in the results of the
activity.
Let us illustrate this. Health is commonly regarded as
a good. Doubtless, there is a state of the materials composing the body that is conditional for health, and is back
of their activity. Bat of that we know, and can know,
nothing. As known by us, health is that state of the
body in which each part accomplishes perfectly its end.
50
GOOD ONLY IN CONSCIOUSNESS.
When the teeth masticate, and the stomach digests, and
the liver secretes, and the blood circulates perfectly, and
every other organ and portion of the body performs perfectly its part, there is health. This state, however, is
itself a form of activity, since a cessation of activity would
be death. As a result of this perfect performance of its
office by each part there is power, and a state of conscious
well-being, in which a person is said to enjoy himself.
Here it must be conceded that the whole worth both of
the state and of the activity, if we choose to distinguish
them, is from their results. If there were from them no
power and no enjoyment, they would be good for nothing. Hiere, what we have to do, and all we have to do, is
to secure that form of activity which we call health. The
results follow by the constitution of God. All that was
said respecting an ultimate end as not lying approximate
to volition applies perfectly here.
And so it is in mind. There may be a state of mind
back of activity, and so back of consciousness, that is good
as related to results; but without those results appearing
in consciousness it cannot be a good. If conceivable at
all in such a state, which I think it is not, mind could
never be known as mind, and, never emerging from it,
would not rise above the dignity of matter.
As there is, then, no good without consciousness, which
involves activity, it would seem that the good must be
found either in the activity itself, or in its results.
But activity in itself cannot be a good. If it had no
results it would be good for nothing, and those results may
be evil and wretchedness, as well as blessing.
We turn, then, in this search, to the results in consciousness, of activity. We are so constituted that any form of
normal activity, physical or mental, produces satisfaction,
51
LECTURES ON MORAL SCIENCE.
enjoyment, blessedness, according to the faculties that act.
Of these the conception is simple and undefinable except
by synonymous terms. They are that in which we rest, in
which we are so satisfied that, within a given sphere, we
look for nothing beyond. From our activity as excited in
taste, by odors, by music, in admiration of beauty, in love,
there may be a satisfaction which shall be the measure of
our capacity in that direction. This all concede to be a
good. We say, then, that in the satisfaction attached by
God to the normal activity of our powers, we find a good,
an end that is wholly for its own sake. We say, too, that
it is only in, and from such activity that we can have the
notion of any satisfaction, enjoyment, blessedness, either
for ourselves or others; and that that form and proportion
of activity which would result in our perfect blessedness
would be right.
Such a form of activity will be to the mind what health
is to the body, and in the maintenance of that will be
found the highest duty and the highest good of man,
his wisdom and his virtue. From it must result to others
all the good he is capable of doing; and to himself all he is
capable of enjoying. Here, as in health, what man has to
do, is to maintain the activity, and God gives the result.
It will appear, from what has been said, that there may
be as many forms of good as there are faculties or forms of
activity; and these forms of good may differ not only in
degree, but in kind. Has man a sensitive nature? There
is, from the activity of that, and from each modification of
that activity, as in the different senses, a sensitive good.
It is multiform, but one. Has he an intellectual nature?
There is from the activity of that an intellectual good.
We may, indeed, conceive of the intellect simply as a
capacity for knowing, and as acting without the slightest
52
THE HIGHEST GOOD.
enjoyment, -as light without heat. But this is not its
actual constitution. Call it what you please, derive it
whence you will, there is enjoyment in the very process
and activity of the mind in the driest mathematical demonstration. Ilas man, again, an Aesthetic, a rational, a
moral, a religious nature? There is, from the activity of
each of these, a corresponding good. It is clear, then, that
the whole good of man would arise from a combination in
the highest possible degree of all these forms of good;
also, that the highest good would be from the activity of
the highest powers in a right relation to their highest
object. Nor is this highest good any mere happening, as
is sometimes said of happiness; nor is it the mere satisfying of any craving; it is that result in God's creatures
that was intended by him, and is an image of his own
rational and holy blessedness in the activity of those
powers in which we are made in his image.
Of the conditions of good the above statement is the
most general that can be made, and admits of no exception. It implies nothing in relation to the direction of
the activity as designed to produce our own good, or that
of others. If there are in man no faculties except for promoting the well-being of the agent himself, then the wellbeing of the agent will be found in the highest activity of
those faculties. But if there are also faculties capable of
working disinterestedly, and that were designed to promote the good of others, then, whatever good can come to
the individual through those faculties, will come from their
highest activity for the very end for which they were
made. - That man, as social, has such faculties, there can
be no doubt; and it may be that it is only in the activity
of these for the good of the whole that the end and highest good of the individual can be found.
5*
53
LECTURES ON MORAL SCIENCE.
In the view just given, we have the basis of a conception of the spiritual universe analogous to that given in
astronomy of the physical universe, but far higher. In
astronomy, no less than in mind, activity, movement, is at
the basis of all order, and beauty, and beneficence. But
there the motion is impressed from without; here it is
firom within; there it is unintelligent; here it comprehends
itself; there it is necessitated; here it is free; there there
is no consciousness and no emotion; here the movement
is reflected in the consciousness, and every faculty sings.
Think, then, of creatures, intelligent, moral, free, with susceptibilities high and keen, in numbers far outnumbering
the starry hosts. See in each a central personality-a
mysterious selfhood, with its attendant faculties revolving,
like satellites, harmoniously about it. See these planetary
intelligences in their myriad heavens, each moving in his
own bright orbit, at once of duty and of freedom, mutually
giving and receiving, and singing together that song which
was typified when the morning stars sang together of old,
-and you have a spectacle which He who sits upon the
central throne may well look upon with complacency, and
pronounce "very good."
That the account now given is correct, appears from this.
If we suppose enjoyment, satisfaction, blessedness, to be
wholly withdrawn from the universe, we should feel, whatever state or form of activity there might be, that its
value was gone. It would be a vast machine producing
nothing. But if we suppose the highest possible blessedness of God and of his universe secured, we are satisfied.
It must surely be difficult to satisfy those who cannot find
an adequate end and good in their own highest blessedness, and in the highest blessedness of God and his universe.
54
THE GROUND OF OBLIGATION.
If the statements now made be correct, we are prepared
to answer the second question mentioned above. The
first was, What ought to be done? and the general answer
w-as, To ascertain the end for which we were made, and to
seek to accomplish that. The second question was, Why
ought it to be done? and the answer is, Because of the
intrinsic excellency and worth of that end. Man, and all
moral beings, are capable, as such, of a high and holy blessedness which can be compared with nothing else, which is
the fruit and crown of all virtuous and holy activity, which
has no exchangeable value, but has, in itself, an infinite
worth.
If it be still asked why a man ought to seek an end
which has this intrinsic worth, the reply is that this idea
of obligation or oughtness is a simple idea, and therefore
that we can only state the occasion on which it arises. Of
its presence in connection with our choice of this end we
can give no account, except that such is our constitution.
This, however, does not compel us to say that we ought
to seek a thing simply because we ought. The sense
of obligation or oughtness may or may not precede the
choice, but could have no place if there were not a ground
of action besides itself. It does not come up out of
vacancy. A man ought to choose that which is congruous
to his nature. It would seem that an act of choice must
be from something in the thing chosen thus congruous.
Hie ought to choose his own well-being rather than the
contrary; but he ought to choose it not simply because he
ought, but because it is well-being. If there were nothing valuable in itself, there would be nothing that ought to
be either chosen or done.
For those who adopt the general line of thought we are
now pursuing, this question concerning good is funda
55
LECTURES ON MORAL SCIENCE.
mental, because there is involved in it the rule for right
action. According to this, any course of action which will
secure the whole good for which man was made must be
right.
But among those who believe that the rule has its basis
in the highest good, there is a difference of opinion as to
what that good is. On this subject I cannot enter at large,
but will refer briefly to two different views. These make
the good consist in that which is conditional for the results,;
and not in the results themselves.
The first is that of Jouffroy. His view is that good
consists in universal order. "When," says he, "reason
first perceives that, as there is a good for us, so is there for
all creatures whatsoever, and that thus the particular good
of each creature is but an element of universal order, of
absolute good, then does the idea of good, so disengaged
and elevated to the sphere of absolute being, appear to
our reason as obligatory."* Hiere two questions may be
asked. The first is, whether the reason does necessarily
form this idea of universal order. Since the reason has
been so much spoken of, nothing has been more common
than to mistake the results of abstraction and generalization for its immediate and necessary ideas. That this is
not one of those ideas, may be inferred from the fact that
men are not agreed in what the order consists. Universal
order may be either that form and extent of activity
which would secure universal blessedness; or that perfect
distribution of good and evil which would constitute moral
order, but would involve punishment and suffering.
But if this idea of universal order be an idea of reason,
it would not follow that the highest good was in that. It
would be only conditional for blessedness. This it doubt
* Introduction to Ethics.
56
THE HIGHEST GOOD.
less is; but if no blessedness were at any time or in any
degree to result from it, it would be in vain. No position
or movement of matter, no activity of mind, however controlled and subordinated, that should have no results beyond itself, would be a good.
These remarks are made on the supposition that the
blessedness is not considered a part of the order. If it be,
then there is simply a confusion of terms. Order would
be made to include not only, according to its usual acceptation, the constitution and movements of the universe,
but its results.
The other view is that of a very able and distinguished
cotemporary. This has its basis in the perfection of the
individual as a moral being, as the other has in that of the
universe as a constituted whole. "The highest good,"
says Dr. Hickok, "the summum bonum, is worthiness of
spiritual approbation." *
From so able a thinker I differ with regret. But what is
that in which a man's worthiness of spiritual approbation
consists? It is in his choice of an ultimate end. The
character is according to that. Does, then, the highest
good of man consist in his choosing as an ultimate end
his own choice of an ultimate end? This cannot be, and
yet would seem to follow firom the definition.
Again, if this be the highest good, it consists of something which can enter into the consciousness but a small
portion of the time, and then only by special effort. Man
can make himself and his state the object of his own
thoughts; but introspection was not intended to be the
business Of his life, nor the form of his activity in which
he should be either most useful or most happy. He was
made to apprehend God and his works, and his fellow-crea
* Moral Science, p. 43.
57
LECTURES ON MORAL SCIENCE.
tures, and to love and admire these, and not to look withiu,
except to correct what may be wrong, and to admire there,
as elsewhere, indications of the divine wisdom and goodness. How, then, can that be the highest good of man
which, if he really had it, he would think of only as the
man who has healthy lungs thinks of his breathing? No
doubt worthiness is conditional, and in a moral being necessarily so, for blessedness. But the word, though it may
be used absolutely, naturally carries with it an indication
of something beyond itself. A worthiness of what? Of
approbation? And why not of the blessedness there is
in and through that worthiness and that approbation?
In this and similar cases the ultimate appeal must be to
consciousness. To that I appeal, only wishing the statements to be so made that the consciousness may apprehend
distinctly the elements with which it is dealing.
In speaking hitherto of activities and their results, language has been used in its ordinary sense, as applied to
outward things. It will be observed, however, that in the
region of mind and of consciousness the results are themselves activities. There is, therefore, a sense in which it
may be said that the activity is the blessedness. The difference is, that what we call activities here are those which
are inaugurated and controlled by the will, while what we
call results are those emotions and feelings which follow
from the other, by the appointment of God. We do not,
therefore, in this connection, regard ends as anything outward, but identify ends and activities.
58
LECTURE III.
KINDS OF GOOD. - SUSCEPTIBILITIES AND POWERS. - GOOD AS HIGHER
AND LOWER. - FORCES AND FACULTIES - THEIR SUBORDINATION. - THE
LAW OF LIMITATION.-METHODS OF ADDITION AND OF DEVELOPMENT. NATURAL AND CHRISTIAN LAW OF SELF-DENIAL.
IN the last lecture two questions were answered. The
first was, What ought man to do? and the second, Why
ought he to do it? Man ought to attain the end for
which he was made; and he ought to do it because of the
intrinsic worth of that end. In answering these questions
we considered the nature of an end as related to rational
activity, and also the nature of good as necessarily included
in an ultimate end.
We now proceed to answer the third question proposed,
which is, How ought man to attain the end for which he
was made? There is a sense in which this question may
be resolved into the first; for, if we know, in the fullest
sense, what to do, we also know how to do it. But convenience and the common use of language justify the
division now made.
In answering the above question we shall naturally examine the different forms of activity of which man is capable, and their resulting forms of good, that we may thus
find for each faculty the law and measure of its activity.
But this may be done with more advantage if we first discriminate between different kinds of good; and if we also
find a criterion by which we may distinguish that which
is higher from that which is lower.
59
LECTURES ON MORAL SCIENCE.
As has been said already, there are as many kinds of
good as there are forms of normal activity; but these
forms of activity may be divided into two great classes
broadly distinguished.
Man has powers, and he has susceptibilities. By his
powers he acts upon external nature; by his susceptibilities external nature acts upon him.
Once awakened, the powers act, not simply because they
are acted upon, but of their own proper activity. The susceptibilities have no activity of their own except as they
are acted upon. In the activity of the susceptibilities the
movement is from without inward; in that of the powers
it is from within outward. In the one we receive; in the
other we give.
When the susceptibilities are acted upon by their appropriate stimuli, the result is pleasure. So far as this term is
employed distinctively, this is the form of enjoyment indicated by it, and is that-which is sought by those who are
called "lovers of pleasure." It has an inlet through each
of the senses. It is the product of warmth, and food, and
of the various kinds of nervous stimulation. That the
production of this is an object in nature, is obvious from
the number and variety of those arrangements by which
sensitive beings receive pleasure from the objects around
them. In this respect the works of God call for our grateful study. Particularly is the human organism admirable
for this in its complex and wonderful adjustment to exter
nal nature.
But in this enjoyment there is no necessary activity of
any rational or moral power. The right relation being es.
tablished, man is no further active than as he has the vital
ity and susceptibility which must be the condition of any
pleasure.
60
SUSCEPTIBILITIES AND POWERS.
Between this form of enjoyment and that from the activity of the powers the differences are radical. And, -
1st. The law of habit, mentioned by Butler, by which
passive impressions become weaker as they are longer continued, applies only to the susceptibilities and the resulting
pleasure. "It is," says Paley, "a law of the machine for
which we know no remedy, that the organs by which we
receive pleasure are blunted and benumbed by being frequently exercised in the same way. There is hardly any
one who has not found the difference between a gratification when new and when familiar, nor any pleasure which
does not become indifferent as it grows habitual." It is,
on the other hand, the law of the powers that they gain
strength by activity, become more masterly, and more
and more capable of being the source of a high joy and
blessedness.
Hlere, then, is a radical contrast between the good from
the susceptibilities and from the powers. The one is like
a vessel flll and sparkling at first, but graduallywasting
away and becoming vapid; the other is like a fountain
whose waters well up the more freely the more they overflow.
A second difference is to be found in the rank of these
two forms of good.
Pleasure is a good in itself, and so an ultimate end; but
for the most part it is also a means to something beyond
itself. This is especially true of legitimate pleasure. It
seems to have been intended as an inducement to the performance of acts which are to have remote consequences
of which the agents themselves are often either ignorant
or regardless. The pleasure of the child, and of the man
too, in eating, and in muscular movement, is the inducement to do that which is necessary for the upbuilding of
61
LECTURES ON MORAL SCIENCE.
the body, but for which they generally have no care. On
the other hand, the good from the activity of the powers,
as in loving and in worshipping, is an end in itself, and has
no reference to anything beyond itself.
There is a third difference. We always feel ourselves at
liberty to forego the enjoyment of pleasure, and respect
ourselves when we do this for the sake of the good which
comes from the activity of the powers, but never the reverse. These two are often, and to some extent naturally,
opposed, and it is a part of the conflict of life to keep
pleasure within its proper limits.
We have thus, from our susceptibilities, a good which
we may call pleasure. From the activity of our powers,
voluntary and moral, we have a good higher and different
in kind, for which we need a distinctive name, but which
we will here call happiness. This will differ with the
powers, intellectual, esthetic, moral, spiritual, which are in
exercise. By these, taking cognizance practically, csthetically, scientifically of the works of God, apprehending the
character and wants of man, being brought into relation
to the attributes and character of God, man is capable
not only of the appropriate enjoyment from such cognitions, but also of putting forth in love all the activity of
his nature for the good of the whole. What of good
there may be from these can be known only by experience, but clearly it need be limited only by our capacity.
My own belief is that that part of our nature through
which we have the highest good lies open to the direct
action of the Spirit of God, as the susceptibilities do to
that of the objects around us; that thus we may apprehend
him directly; and that in his response to this, in love, man
is capable of a good that is ineffable, and may be called
"fulness of joy," or blessedness. The capacity for this I
62
FORCES AND FACULTIES.
suppose as much belonged to man originally as his capa
city for perceiving beauty.
The above distinctions are practical, and, from the tendency there is in men to seek pleasure in opposition to
their higher good, are worthy of careful attention.
We now turn from this broad classification of good to
inquire for the basis of one that is more exact. We speak
of good as higher and lower, and we have an instinctive
feeling that some forms of good are higher than others.
Is there a criterion by which we may determine what is
higher and what is lower?
In answering this question, I hope for indulgence if I
enter upon a range somewhat wide. Moral science has
usually been studied as isolated. My wish is to connect it
with the laws of that physical system which not only supports man, but has its culmination in him. I wish to show
that there runs through both one principle of gradation,
and one law for the limitation of forces and activities,
and so of the forms of good resulting from them. If this
can be done, it will add to both physical and moral science
the beauty of a higher unity than has commonly been
noticed, and will show that there could have been but one
author for both.
All good, and all arrangements conditional for good, are
the result of some activity. They are in or from it. Arrangements conditional for good are the product of forces,
good itself of faculties. A faculty is a force united to personality and subject to the control of the will. What we
need to find, then, is a common law for the subordination
and limitation of both forces and faculties.
This we find in their relation to each other as conditional and conditioned. The forces that are at work
63
LECTURES ON MORAL SCIENCE.
around us and the faculties within us, from the lowest to
the highest, may be ranked as higher and lower as they
are or are not a condition one for another. That which is
a condition for another is always the lower.
In anything we may choose to examine, -a house, or
a portion of matter, -we shall find some conceptions or
properties that may be spared, and yet the thing continue
to be that thing. But we may continue our analysis till
we reach certain properties or conceptions which are indispensable, which underlie all others, and are conditional for
that thing. So it is with solidity, or the occupation of
space, in matter; so with the foundation of a house.
These may be of no importance in themselves, but allimportant as conditions for something above them.
It is this relation of the forces of the universe and of
their products to each other, as conditional and conditioned, that gives to it its unity. If its forces were
diverse, it would not be a universe, -that is, if they were
so diverse as to be free from this relation. Any being or
thing conditioned upon nothing in the present system, and
the condition of nothing, would be so utterly out of relation as to be alien from every conception of unity.
In seeking, then, for the law of subordination and limitation of the forces of the universe, we must begin at the
lowest, and to find that, we must continue to drop from our
conceptions of the universe every force and product that
can be spared till we reach that which being taken away
the universe would be dissipated, would become utter
chaos, and so, having no unity, would cease to be a universe. What is that force? Plainly it is the law of gravitation. By this, particles of matter that would otherwise
be chaotic, are aggregated, and its masses move in harmony. This is a universal force. It is conditional for the
64
HIGHER AND LOWER FORCES.
activity of every other,.and is the lowest of all. The product of this would be mere unsorted matter aggregated
and moving in systems, and would be the lowest conception we could form of a physical universe. It would be
the first approximation towards a good, - the first step
conditional for all others; for that which we find last in
thus going back must have been first in the order of nature,
if not of time.
Gravitation being thus given, what, in going down, is
the last force we should have been obliged to drop before
reaching this? What, in going up, would be the next
step to fit matter for any use to which we can suppose it
might be put? It would be to bring matter, chiefly of
the same kind, into solid masses by what we call the
attraction of cohesion. For this gravitation is plainly conditional, since matter must be aggregated before it can
cohere. This gives us the next higher force.
The next force needed, for we will now pass up, is chemical affinity. By this, particles of matter having different
properties are united, and form compounds. In the present state of our knowledge it cannot, perhaps, be proved
that cohesion is always conditional for chemical affinity.
If not, these two forces must be ranked with those groups
to be spoken of hereafter. The compounds, however,
formed bythis force are conditional for the action of that
power which we call life. The power of life assimilates
nothing which has not previously entered into combination
by this affinity.
Through the action of the three forces now mentioned
we may have the conception of a world, inorganic, destitute of life, and having its unity solely firom the fact that
its forces are thus conditional and conditioned.
But the inorganic world is conditional for that which
6*
65
LECTURES ON MORAL SCIENCE.
is organic, and is under the control of that principle or
force called life. And here, again, we have three great
forces with their products. These are the vegetable, the
animal, and the rational life.
Of these, vegetable life is the lowest. Its products are
as strictly conditional for animal life as chemical affinity
is for vegetable life, for the animal is nourished by nothing
that has not been previously elaborated by the vegetable.
"The profit of the earth is for all; the king himself is
served by the field."
Again, we have the animal and sensitive life, capable of
enjoyment and suffering, and having the instincts necessary to its preservation. This, as man is now constituted,
is conditional for his rational life. The rational life has its
roots in that, and manifests itself only through the organization which that builds up.
We have, then, finally, and highest of all, this rational
and moral life, by which man is made in the image of God.
In man, as thus constituted, we first find a being who is
capable of choosing his own end; or, rather, of choosing
or rejecting the end indicated by his whole nature. This
is moral freedom, and in this is the precise point of transition from all that is below to that which is highest. For
everything below man the end is necessitated. Whatever
choice there may be in the agency of animals of means for
the attainment of their end, - and they have one somewhat wide,- they have none in respect to the end itself.
This, for our purpose, and for all purposes, is the characteristic distinction, so long sought, between man and the brute.
Man determines his own end; the end of the brute is
necessitated. Up to man everything is driven to its end
by a force working from without, or from behind; but for
66
SUCCESSIVE PLATFORMS.
him the pillar of cloud and of fire puts itself in front, and
he follows it or not, as he chooses.
In the above cases it will be seen that the process is one
of the addition of new forces, with a constant limitation
of the field within which the forces act. The sphere of
gravitation is wider than that of cohesion. Cohesion rests
upon it as upon a base. The sphere of cohesion is wider
than that of chemical affinity; that of chemical affinity
wider than that of life; that of vegetable life wider than
that of animal life; and that of animal life wider than
that of rational life. Hence, the plan of the creation may
be compared to a pyramid, growing narrower by successive platforms. It is to be noticed, however, that while
the field of each added and superior force is narrowed, yet
nothing is dropped. Each lower force shoots through, and
combines itself with all that is higher. Because he is
rational, man is not the less subject to gravitation, and
cohesion, and chemical affinity. HIe has also the organic
life that belongs to the plant, and the sensitive and instinctive life that belongs to the animal. In him none of these
are dropped; but the rational life is united with and superinduced upon all these, so that man is not only a microcosm, but is the natural head and ruler of the world. IHe
partakes of all that is below him, and becomes man by the
addition of something higher.
If now we pass to the physical system of man, we shall
find that it is composed of various systems and groups of
systems which are conditional and conditioned in the
same way.
Here again there are three divisions. In the lowest
group we have those systems which are for building and
repairing; in the next higher, those which are for support
and locomotion; and in the third those which are for sensea
LECTURES ON MORAL SCIENCE.
tion and direction; and each lower group is conditional for
the higher.
In the several groups, also, the same general order holds.
Among the builders or repairers the nutritive or digestive
system is the lowest. This is conditional for the circulatory, this for the respiratory, and this again for the secretory and assimilative. In the systems for support and
locomotion, the osseous system is conditional for the muscular; and that system of nerves which is for sensation is
conditional for that which is for motion and direction.
Whether these subordinate systems can all be placed
in a right line is not important. It is now conceded that
in the classification of animals and of plants there are
groups within which no precise order can yet be traced.
But in all cases, - and here is the principle contended
for, -if the end accomplished by any system or group be
conditional for any other end beyond itself, it will be lower
than that end. Thus, building and repairing are lower
than support and movement; and these are lower than
sensation and direction.
Nor does this law stop here. It applies to the mind.
In this, too, according to the latest and best classification,
there are three groups, and each lower is conditional for
the higher. There is first the intellect, including what are
sometimes called the cognitive faculties,- all our faculties of knowing. These are conditional for the emotive
or pathematic nature, including all the feelings and emotions consequent upon knowledge. These again are conditional f6r what Sir William Hamilton calls our conative
powers, those of desire and of will.
In each of these we have a group, which we need not
now examine; but we shall find running through each the
same principle of order and arrangement already noticed.
68
METHOD OF DEVELOPMENT.
We have thus a beautiful gradation from those "foundations of the earth" laid by God, and "the corner-stone
thereof," up to the point at which "the morning stars sang
together, and all the sons of God shouted for joy."
But in attaining and preserving the unity and order of
the universe, God's methods are two. Besides this of addition, there is another applicable only to organic beings,
that of development. In all organic beings there is something central and enveloped, and the being reaches his perfection by being developed.
In some respects this is the reverse of the other method.
In that, in making our analysis, and seeing what we can
spare, we reach that which is lowest; but in this, by the
same process, we reach that which is highest. If we ask
what the last thing is in the universe that can be spared,
and unitv remain, it is gravitation, the lowest force; but if
we ask what the last thing in man is that can be spared
and he remain a man, it will be that in him by which he
is highest. In the method of additions that which is most
fundamental, which is first in the order of our conceptions,
is lower than that which is later, and serves it. But in the
method of development that which is the most fundamental and first is the highest, and all else is lower as it
is less or more essential to this. Here the lower are a
condition for the development of the higher, but still are
conceived of as coming in later. Here, therefore, when
anything is spoken of as a condition, it is not to be regarded as a condition of being, but of development. In
both methods the principle of arrangement already stated
will hold; that is, if the end accomplished be a condition
for any other end beyond itself, then it will be lower than
that end, and all the means and apparatus for producing
it will be lower than those for producing the higher end.
69
LECTURES ON MORAL SCIENCE.
In the range of conditioned forces and systems above
spoken of we find no good till we come to the gratification
there is in the lowest sensitive being from the assimilation
of food, and in the performance of those functions which
are at once the condition for life, and by which life manifests itself. From that point the rank of the good rises
precisely as the systems do through their whole gradation
till we reach the highest of all.
We thus find the law of subordination both of forces
and their products, and of susceptibilities and faculties,
and of the good resulting firom their activity. This we
needed here because there is involved in it, or results
immediately from it, what I shall venture to call the law of
limitation. By this I mean the law which fixes the proper
limit of every form of activity, and so of every kind of
good except the highest; and so will enable us to live in
the best sense of that much-abused expression, "according
to nature."
This is a point of great importance in morals. According to an ancient theory, that of Aristotle, virtue and good
consist in proportion, or the golden mean. It is readily
seen that many things, that most things which men use
and enjoy, are good up to a certain point, but that, carried
beyond that point, they become, if not in themselves, yet
relatively, evil. The pleasures of the senses and of the
appetites are good, but may be readily carried to excess. -
Where is the limit? Amiusement is good, the pursuit of
money is good. Where is the limit? There is a wide
range of questions which arise at this point in respect to
the use of things lawful. How far may we go in dress?
in expense? in conformity to fashion, and the usages of
those around us? To determine these questions we need
some plain criterion. Besides, there are those who think
70
THE LAW OF LIMITATION.
all pleasure and good alike except in intensity and duration. Paley thought so. The sensualist makes an irruption upon us and says that his joys are as high as any
others, -that is, in his opinion, and that, on such a subject, the opinion of one man is as good as that of another.
It is a mere question of taste and feeling, and there is no
standard. We are also asked by another class, as by
Whewell, "How are we to measure happiness, and thus
to proceed to ascertain by what acts it may be increased?
If we can do this, then indeed we may extract rules and
results from the maxim that we are to increase our own
and others' happiness; but without this step," which he
plainly supposes cannot be taken, we "can draw no consequences from the maxim."* For such cases and inquiries
we need a law of subordination and of limitation, a test
and measure both of activity and of good.
If man would enjoy his whole good, it is obvious that
his life must be a unity as the universe is, so that all the
forces that conspire to make it up may act in harmony.
This would give all possible good. But the method of
attaining this is clearly set before us in the method pursued by God in making the universe one. As the forces
in man, that is, his faculties, bear the same relation to each
other that the forces in nature do, we shall find their
proper limit by finding the limit which God fixes in proportioning the conditional and conditioned forces of nature. His method of doing this is to give to each lower
force precisely the relative strength that shall make it
most perfect as a condition for the activity of those above
it, and which are conditioned by it. It is to carry that
which is an end in one sphere no further than will fit it
to become a means for the one next above. Gravitation
* Vol. I., B. III.
71
LECTURES ON MORAL SCIENCE.
is a necessary means and condition of cohesion; these two
of chemical affinity; these three of organic life, and so on;
but no lower end is pursued one jot beyond the point
where it may become the means of a higher end. Each
force is limited at the point where it may best subserve
the force above it. If gravitation were stronger than it
is it would prevent the ascent of the sap altogether, or
would cause a dwarfed vegetation. Man could not lift his
foot, or only as he now pulls it from a clay-bed. The bird
could fly but a short distance, and with weary wing. If
its force were less, we should be liable to be blown away,
the equilibrium of all bodies would be less stable, and the
earth would be constantly sending off meteoric stones to
astonish the inhabitants and puzzle the philosophers of
other planets. As it is, gravitation will be found to have
precisely that force which is best for the stable equilibrium
of bodies, for the ascent of sap to its proper height, and in
animals for agility and firmness combined. If chemical
affinities were stronger than they are, the power of life
would be unable to disengage the materials with which
to build up the plant and the body; if they were weaker,
that power could not prevent vegetable and animal decomposition and corruption even before death. Thus shall
we find it throughout the whole range of forces in nature
and in man. Hence the law of limitation will be, that
every activity may be put forth, and so every good be
enjoyed, up to the point where it is most perfectly conditional for a higher good. Anything beyond that will be
excess and evil. It is a peculiarity of the works of nature
as distinguished from those of men that her ends are also
means, but she never pursues such an end beyond the
point at which it would cease to be a means.
Here, then, is our model and law. Have we a lower
72'
THE LAW OF LIITATION.
sensitive and animal nature? Let that nature be cherished and expanded by all its innocent and legitimate enjoyments, for it is an end. But, - and here we find the
limit,-let it be cherished only as subservient to the higher
intellectual life, for it is also a means. $Let the intellectual
nature have its full growth; let it scale every height, and
sound every depth, for it is an end; but let it do this only
in subservience to the higher emotive, moral and spiritual
nature, for this, too, is a means. Thus let each of these,
while it fulfils its own ends, so fulfil them as to minister
to the sphere above, until we come to that which is not
a means, but is of itself an end, and an absolute good.
Men may enjoy pleasure, may use intoxicating drinks and
narcotics to any extent they please, provided it shall interfere with no higher good. They may indulge in expense,
amusements, fashion, as they will, if there is nothing higher
and better that they can do. Certainly if there is nothing
better they can do, they had better do that. The law applies universally so long as there is a good that is conditional for one above it, - so long as there is an end that
is also a means. But when we reach the highest and supreme good, as that is conditional for nothing beyond
itself, there can then be no excess. That is infinite; it
is the ocean without a bottom or a shore.
Up to this point this system has fully met the wants of
that part of our nature whose activities have a natural
limit which cannot be passed without degradation and
loss on the whole. At this point it meets those indefinite yearnings which testify to the connection of man with
the Infinite, and are the presage of his immortality.
We may now readily see how far Aristotle was right.
His system had a basis, and not a narrow one. Much of
our good is the result of proportion and limitation, and
73
7
LECTURES ON MORAL SCIENCE.
of finding the golden mean. He was right so far as he
went, but he needed the law of limitation, and he did not
see the ocean.
It will be observed that none but a good man can adopt
the model above proposed, for no bad act can be at once
an end and a means. Lying, cheating, stealing, are means
only, and can never become ends; but every good act is
not only an end in itself, but is also a means of confirming
him who does it in habits of goodness; and thus he who
adopts this model will find provision in it that his path
shall be as that of the just, "shining more and more unto
the perfect day."
The law of limitation, above given, implies the natural
law of self-denial.
This requires us to reject no good cynically because it is
a good. It respects every part of the human constitution
as made by God, and gives free play to every activity
within its own limits. It says, with an apostle, that
"every creature of God is good, and nothing is to be
refused if it be received with thanksgiving." Any supposable strength in the appetites will only give force to the
character, provided the governing powers keep them wholly
under control. No matter how strong and spirited the
horses if they are trained to perfect subjection. So with
the desires. The desire of knowledge and of power and
of esteem cannot be too great if they do not conflict with
the affections and the moral nature. As they are stronger,
they will but afford a richer soil in which these can strike
their roots, and thus furnish the sap for a more abundant
fiuitage. And so it is with every lower form of activity.
The stronger it is, the better for those above it, if it does
not conflict with them. The stronger and more healthy
the body, if a man be not at all animalized through it, the
74
LAW OF SELF-DENIAL.
better for every mental faculty, and for every high and
healthful form of affection and emotion. The law requires
the restriction or denial of every appetite, desire, propensity, passion, at the point where it would interfere with
something higher, and only at that point. This is the natural and original law. But if moral disorder has come
in and become habitual, if great interests are at stake in
circumstances of temptation and struggle, it may be wise,
and even a duty, to ignore and reject many pleasures that
might otherwise be indulged in, as the soldier who hastens
to defend his country may not stop to enjoy fine scenery
by the way.
This gives us the difference between the natural law of
self-denial and the Christian law. The first would be the
law for a man in health, simply requiring that nothing
should be done to injure that. But Christianity is wholly
a remedial system. "They that are whole need not a physician, but they that are sick;" and the law of self-denial as
a remedy, or as a condition for the working of other remedies, may be as different from its natural law as the regimen of a sick man should be from that of one who is well.
It has been from a consciousness of disorder that difficulties and obscurity have arisen at this point. There has
been a feeling that self-denial, as well as self-torture, was
compensatory; and then, when the lower powers had gone
to excess, it is not strange that there should be a tendency
to their undue repression, and even eradication. This
has given rise to asceticism, and penances, and to a vast
brood of superstitious observances. But precisely what
the natural law is in its place, that the Christian law is in
its place. Under Christianity self-denial is not a remedy,
but the condition for the working of remedies, and its law
75
LECTURES ON MORAL SCIENCE.
is that it shall be carried just so far as is necessary for the
best working of those great remedies which God has provided for the moral disorders of this world. This may
often make self-denial very severe, but only as it is salutary.
It may require the cutting off of a right hand, or the plucking out of a right eye, but only on the condition that they
"offend," that is, cause you to stumble in your course
towards heaven.
In what has been said hitherto, the dependence of the
higher upon the lower forces and powers has been prominent. So long as these powers remain within the limits of
unconsciousness, the right proportion is always preserved;
but when they come under the direction of a finite, and
especially of a perverted will, that proportion is not preserved. The danger is that the dependence of the higher
upon the lower will be ignored, that the lower will in
consequence be neglected and deteriorate, and then that
the higher itself, the fountain of its sap being dried, will
dwindle and wither. So is it always when a short-sighted
selfishness would snatch too soon and grasp too much; so
always when men would reach their ends by circumventing or evading those laws by which God has appointed
that they should be gained.
The law - and this is especially true in organic life - is,
that that which is highest can increase only through the
ministration of the parts that are lower, and hence that the
perfection of the highest in its sphere can be reached only
as the lower are made perfect in their sphere. In training
a child, would any one secure the highest, the best balanced,
and the longest continued action of the mind, he can do it
only by so attending to the body as to secure the priceless
but subordinate blessings of health and a sound physical
76
METHOD APPLIED.
constitution. Would you have healthy feeling? Cultivate
the intellect, else feeling will be fanatical. So has God
constituted every organic being that "if one member suffer all the members suffer with it." Yea, and so that
upon "those members of the body which we think to be
less honorable we should bestow more abundant honor;"
since the perfection of the more honorable members that
are ministered unto can be attained only through the perfection of the less honorable that minister. Our end may
be the perfection of the higher; our method must be to
secure it through the perfection of the lower.
This method is one of wide application. It teaches us,
while we aim at the highest, to care for the lowest; while
we aim at the mind, to care for the body; while we aim at
a perfect government, to care for the people and to seek
to educate and elevate them; while we aim at perfect
social organizations, to give woman her true place, not as
inferior, but as different. No element of reaction upon
progress can be swifter or more fatal than that of degraded
mothers. It teaches us to care for children, and servants, and slaves, and criminals. Nature herself seems to
cry out to us to do this. All history shows how men
have disregarded this method and law, and it shows, too,
how the law has avenged itself by bringing down the high
and the low together. This is indeed the one great lesson
of history. It needs to be pondered, more especially by
republics, where the barriers of form and of force are so
feeble; but whatever the form of government may be, the
law is as pervading and resistless as that of gravitation,
and the result is only a question of time. That result no
form of heathen civilization has been able to prevent. It
can be but one so long as successful men and successful
7*
77
LECTURES ON MIORAL SCIENCE.
classes seek with a blind selfishness to elevate themselves
at the expense of others, -so long as men refuse to adopt
the models of method which God has set before them, and
thus to bind society together in an organic and a perfect
whole.
so
LEOTURE IV.
RELATION OF INTELLECTUAL AND MORAL PHILOSOPHY. - SPONTANEOUS
AND VOLUNTARY ACTIVITY. - FACULTIES INSTRUMENTAL AND ULTI MATE. - INSTINCT. - TIHE APPETITES. - NATURAL - ARTIFICIAL. - THE
DESIRES. - CLASSIFICATION OF THEI.- DESIRE OF CONlTINUED EXIST ENCE.
THEp nature and limitations of good having been already
discussed, we now proceed to consider those powers from
the activity of which good results.
This brings us to that point both of union and of cleavage between mental and moral science, at which, as we
have seen, no little confusion has arisen. Theoretically
the line between them is, or may be made, distinct; but
practically the treatment of the one will include, in some
measure, that of the other. What man ought to do will
depend on what he is, and the circumstances in which he
is placed. Mental science, or psychology, will, therefore,
be conditional for moral science, whic h will make use of
the first, and is the higher of the two. The province of
psychology will then be to show what the faculties are;
that of moral philosophy to show how they are to be used
for the attainment of their end. Both have to do with the
faculuties of the mind, but in different aspects; as both the
botanist and the agriculturist have to do with wheat, and
the astronomer and navigator with the heavenly bodies.
The botanist classifies wheat; the agriculturist raises it,
and cares for a knowledge of its class only as it will ena
79
LECTURES ON MORAL SCIENCE.
ble him to do that. The astronomer investigates the nature of the heavenly bodies and their relations to each
other; the navigator regards them solely as the means by
which his course may be guided. And so the moral philosopher does not care for the nature and classification
of the mental faculties except as a knowledge of these
will guide him to their right use and proper end. So far,
however, as this knowledge will thus guide him, as to a
great extent it will, he is bound to have it.
The moral philosopher is, therefore, not excluded from
the domain of the psychologist. It is his domain. It is
the soil into which his science strikes its roots; it is indispensable for him that certain portions of it, at least, should
be rightly cultivated; and if the psychologist does not
do his work in those portions as he thinks it ought to be
done, he has a right to revise it, and do it for himself. It
is not to be allowed that the mere psychologist may lay
down such doctrines as he pleases respecting the moral
nature, and thus virtually determine the character of the
science. It will, moreover, always be necessary to consider the faculty itself in determining its use, and to
make our classifications with reference to the objects of
moral science.
In accordance with this we shall, -
I. Distinguish the two great forms of mental activity.
These are, -1st. The Spontaneous. 2d. The Volhntary.
And, II. We shall class the mental faculties as they are re
lated to ends. And,
1st. Of the mental activities, as they are either, 1st,
spontaneous; or, 2d, voluntary.
As the inorganic world underlies and is conditional for
the vegetable world; as the vegetable is conditional for
80
LIFE AUTOMATIC.
the animal world; as the automatic or organic life of the
body is conditional for its animal life, so is there an automatic and involuntary life of the mind that is conditional
for its voluntary and responsible movements.
All life is inscrutable, and to our view automatic. Hiow it
begins it is impossible for us to conceive, since it manifests
itself only through organization, while there is no organization that is not its product. In vegetables its results are
seen in organizations entirely destitute of sensation and of
will; and in the animals and in man, the conditions being
complied with, it works with the same independence. The
circulation of the blood in man, digestion, secretion, assimilation, have organs appropriated to them which the will
does not reach, and they go on by laws as independent
of the will as the circulation of the sap in vegetables.
Through these organs and processes there are built up
and-presented to us the organs of sensation and of voluntary motion, but we cannot say what they shall be.
We cannot cause this power of life to build up such a
structure as we should like; we cannot add one cubit to
our stature, or make one hair white or black.
But precisely as we find the heart beating, and accept
the limbs already built up, so do we find the mind thinking, and the faculties acting, and accept them as they are
given. Those cravings which we call appetite are upon
man from no contrivance of his. He knows and can know
them only as he finds them acting. HIe finds a succession
of thoughts bubbling up, like water from a fountain, of
which he knows not tie source, and the flow of which he
can no more stop than he can the flow of a river. No man
ever thought at first by willing to think. Adam did not.
He was created a thinking being, and thought as naturally
and as necessarily as he breathed. Nor can any man stop
81
LE CTURES ON MORAL SCIENCE.
thinking by willing it. Hie must think. HIe may control
the current of his thoughts, but think he must; and if
his thoughts had flowed on forever, as they do in dreams,
without the intervention of a personal power, he would
have been a thinking thing. Man, also, feels desires
springing up. These he may or may not gratify, but there
they are, a part of his nature. The natural affections, too,
put forth their tendrils like the vine, and quite as independently of any will of man.
With these faculties the self-conscious, rational, personal
being, with powers of supervision and comprehension, is
endowed; into this nature is put, or rather we may say is
so incorporated with it that it becomes a part of himself.
This nature is an epitome of all that is below him, and he
was put into it not only that he might govern himself, but
govern it, as we saw in the last lecture, after the model of
that government which God exercises over nature itself.
This is the garden into which man is put that he may
dress it and keep it.
Am I, then, distinctly understood at this point? Is it
seen that there are activities going on within, not only our
bodies, but our minds, with which our wills have as little
to do as with the springing up of the grass? These faculties and activities are one thing, and we are another. We
are responsible for the activities only as we can control
them directly or indirectly.
In this original and spontaneous nature there are char-l
acteristics common to all men, and also diversities apparently as great as in natural scenery. Some natures are
richer and grander than others; they tower up like the
great mountains. Some are more easy of control, and
some more difficult.
We now proceed, as was proposed, to the consideration
82
CLASSIFICATION.
and classification of our various faculties and powers as
they are related to ends.
In this aspect the faculties or powers may be divided
into two great classes: -
I. Those which are instrumental for the attainment of
ends beyond themselves. This is the first class. Here we
find, -
1st. Those which indicate ends. These are the Instincts, the Appetites, the Desires, and the Natural Affections. And, -
2d. The Intellect, in the light of which we pursue ends.
These are the Instrumental Powers, and do not necessarily
imply a moral nature. They require to be governed.
II. The second great class of powers are those in whose
activity we find ends beyond which there are no others.
These are our Moral Nature. By them we elect and sanction ends. They govern, or, at least, ought to govern.
These are the powers that belong to man as a person.
They are Reason, Moral Affections, and Free-will.
The Instrumental Powers are neither good nor bad in
themselves, but as they are used. Generically we share
them with the animals, but they are much modified by
being taken into connection with a higher nature.
Let us, then, first consider those powers which indicate
ends.
In the conception of an end the primary element is not
intellectual. If there were no original, no rational apprehension of good involving desirableness, congruity, automatic tendency, impulse, appetency or craving, revealing
some want to be satisfied, or capacity of enjoyment to be
met, we could have no conception of an end. In our analysis-in this direction this is the last thing that we reach,
and so is conditional for all the rest. The intellect is im
83
LECTURES ON MORAL SCIENCE.
plied. There must be consciousness. Every mental operation, whether perceptive or impulsive, must take place in
the light of that. But consciousness being given, the impulse towards an end or the apprehension of it as having
in it a good, is the primary element in our conception of an
active, as distinguished from a contemplative being. Without such impulse or apprehension, the objects we now seek
might be known as they are in themselves, but not as ends
for us. There would be no motive for the voluntary exertion of the intellect even. As a part of our nature, these
impulses are generically the same in all men, but reveal
themselves in different proportions, and in them we find
what have been called the active powers of man. By
this it is not meant that the contemplative powers are not
active, but that they do not, and these do, lead to action.
The powers which indicate ends are commonly, and, as
it seems to me, correctly divided into the Instincts, the
Appetites, the Desires, and the Affections. Of these there
is no question respecting any except instinct, the existence
of which in man has sometimes been doubted.
Instinct, which we shall first consider, is defined by
Paley to be "a propensity prior to experience, and independent of instruction." It leads animals obviously destitute of either understanding or reason to perform the
same acts as if possessed of those powers in the highest
degree. In building her cells the bee proceeds on the
principles of mechanics and of the abstruser mathematics.
In incubation the hen seems to have a knowledge of the
doctrine of different specific gravities, and turns her eggs
over regularly because the yolk is slightly heavier than
the white. Animals with migratory, and those with acqutisitive instincts, proceed on an apparent knowledge of
the movements of the heavenly bodies for months in the
future.
84
INSTINCT.
In all animals of the same species instinct is mostly uniform, and, as we descend in the scale of creation, becomes,
in the inverse ratio of understanding and reason, more
uniform, more blind, and more perfect. A pure and unperverted instinct may always be trusted implicitly. A
marvellous and a beautiful thing it is to see "the stork in
the heaven knowing her appointed times; and the turtle,
and the crane, and the swallow, observe the time of their
coming." Surely, here "lie leadeth the blind in a way
that they know not." hIere extremes meet -the perfection of reason and the perfection of ignorance.
But as the light of understanding and reason increases,
the glimmerings of instinct seem lost. Accordingly, most
writers on morals have not noticed this as one of the active
powers, or, if they hlave, have spoken of it as confined
almnost wholly to animals. But if instinct is needed by
rational creatures we shall be sure to find it, for God does
not care less for them than for the ant and the bee. It
would be in accordance with all we have hitherto seen of
the order of the universe, and of the mode in which its
unity is secured, if we should find this, like gravitation,
passing up and blending itself with the activity of the
very highest power of its own order. Or, if any should
suppose that this, the lowest form of intelligent action,
cannot blend with those intuitions of reason which it so
much resembles, it is yet pleasing to see in its certain
guidance the best analogon and symbol of perfect reason,
just as gravitation, which is the lowest motive power, is
the best symbol of love, which is the highest of all.
I suppose, however, that something of instinct does
blend with the activity of our highest powers. For this,
it is not necessary that we should be under the guidance
of ainy specific instinct, for wherever there is a tendency
8
85
LECTURES ON MORAL SCIENCE.
in our nature that is automatic, there we find the instinctive element. Hence we may, and do, speak of rational
instincts. In every created nature, however high, there
must be tendencies and yearnings by which the true end
of the being shall be revealed to itself, and in which the
first movements towards that end shall originate. That a
good of any kind should begin to be sought in any other
way, is not conceivable. And so the Scriptures represent
it. They speak of thirsting for God; and the Saviour said,
"If any man thirst, let him come to me and drink."
Our associations with instinct may be low; but it is
really a high and sacred thing. In it we see the Highest
stooping to the lowest, and illustrating that care and guidance of which they may feel secure who follow the promptings of any nature that is unperverted, and as it came
from his hand.
We now proceed to the Appetites. These are those
cravings of the animal nature which have for their object
the well-being of the body and the continuance of the race.
These are to be distinguished from a desire for those
pleasures of the palate, for example, with which they
become so intimately associated that they are seldom
thought of separately. The craving is purely instinctive,
and, as such, has in a healthy state the infallibility of instinct, both in indicating and measuring the wants of the
system; but the pleasure of eating and drinking will be
according to the quality and condiments of the material
taken. This pleasure may be perpetuated far beyond the
point at which the craving is satisfied; and the modes
of causing it may be reduced to a system and a science.
The science of cookery will be useful as it fits substances to
satisfy the craving, and so for assimilation; it will be inju. rious as it merely stimulates the palate. If the substance
86
THE APPETITES.
stimulate the palate slightly, or not at all, as water, the
craving is simply satisfied, and there is no danger of excess; but the more stimulating the substance, either to the
specific sense connected with the appetite, or to the nervous
system generally, the more danger there is of excess from
confounding the excitement of the sense, or the nerves,
with the demand of the system.
According to Stewart, the appetites are distinguished
by three circumstances. 1st. They take their rise from the
body. 2d. They are periodical. 3d. They originally imply an uneasy sensation, afterwards, upon experience, a
desire for their appropriate objects.
The appetites are usually said to be three,- hunger,
thirst, and the appetite of sex. But there are tendencies
and cravings that may more properly be classed with the
appetites than elsewhere. These are the craving for air,
for exercise, for rest, and for sleep. These all take their
rise from the body, are periodical, and originally imply an
uneasy sensation; afterwards, upon experience, a desire of
their appropriate objects. They also require to be regulated on precisely the same principles as those commonly
ranked as appetites; and it may be well to place them
here, as bringing them nearer the conscience, since all concede that the regulation of the appetites is a duty.
The necessity of the appetites for the accomplishment
of their immediate ends is well stated by Reid. "Though
a man knew," says he, "that his life must be supported by
eating, reason could not direct him when to eat, or what,
how much, or how often. In all these things appetite is a
much better guide than reason. Were reason only to
direct us in this matter, its calm voice would often be
drowned in the hurry of business or the charms of amusement. But the voice of appetite rises gradually, and at
87
LECTURES ON MDIORAL SCIENCE.
last becomes loud enough to call off our attention from
any other employment."
As they are means of sustaining the body and continu ing the race, the appetites are the condition for all that is
above them. But besides the direct objects thus immedi ately secured, they are also closely related to industry and
the social affections. The craving, which is the radical
and constant element in the appetite, is related to industry,
and the pleasure, the incidental and variable element, is
related to the social affections.
When we observe how busy a scene this world is, and
what human labor has accomplished, -the forests it has
cleared, the fields it has cultivated, the cities it has built,
the ships it has constructed, the oceans it has navigated,we are little apt to think how much of all this is owing to
so simple a cause as the appetite of hurger. "All the labor
of man," says Solomon, "is for the mouth, yet the appetite
is not satisfied." Food is our first, and is a constantly recurring want; and probably the amount of labor for obtaining and preparing it is greater than for all other purposes.
When the savage has plenty of food he does little but eat
and sleep, and only the stimulus of hunger can goad him
on to the labors of the chase. In civilized commnunities,
those who turn the soil, and hew the wood, and lay the
brick and mortar, are generally those who labor for their
bread; nor is it probable that a less imperious motive
would induce the effort. Nor is it bodily activity alone
that is excited by this stimulus. Hunger, rather than any
of the nine, has been the muse of some of the best poets.
This connection of the appetites with industry, which is
so indispensable to force of character and to all good habits, shows that they were intended by God to be ministers
of human virtue, and not the occasions of vice.
88
THE APPETITES.
But the appetites are also connected with the affections.
So naturally do our kind feelings rest upon those who
share the same table with us that "to eat bread" with one,
that is, to receive or furnish hospitality, has been regarded
in many countries as a pledge of kindness and good faith.
"He," says the Scripture, as if it aggravated the treachery,
-"he that did eat bread with me hath lifted up his heel
against me." It was from the connection of the appetites
with the social feelings that the drinking customs of society derived much of their power and also of their danger.
It was the social glass that led young men of generous
affections to occasional excess, and the appetite was then
cherished and justified on the ground of indulging the
social nature, till the capacity for social enjoyment was
diminished, and the man sunk into degrading habits of selfish, solitary, animal gratification.
It is from this natural and intended connection of the
affections and virtues with the appetites that we are not
degraded by them. We share them, indeed, in common
with the brutes; but they so underlie our higher nature
and maypo blend with it as to become the occasions of
some of its most beautiful manifestations, and when confined within the bounds of reason and religion are the
occasion only of good. The man who eats that he may
live and improve his higher faculties, and do good, is a
man. But the. man who lives that he may eat is a brute.
A course of indulgence of the appetites has been called
a life of pleasure. But retribution reaches to the body,
and there could be no greater misnomer. Every excess is
sure to be punished. Besides the penalties of immediate
reaction and specific disease, by the law of habit already
noticed, the capacity for enjoyment becomes gradually
89
8*
LECTURES ON MIORAL SCIENCE.
less, and no object is more pitiable than a man who is beginning to taste the dregs of such a life.
At no point do the dictates of virtue and of an enlightened self-love more clearly coincide than in the regulation
of the appetites. The proper notion of temperance with
reference to them is not an abstinence from any particular
thing, but such a control of all the appetites as will result
in the greatest power and activity both of body and of
mind, and as shall subject them most fully to our control.
Anything short of this is criminal, and infallibly pernicious;
and any use or enjoyment of the appetites compatible with
this may be allowed.
From the above account it is most plain that the law of
the appetites is to be found in their encl. That end we
have the capacity to see. We can also see the fitness of
the appetites for its accomplishment, so that when we
yield ourselves to the guidance of an unperverted appetite we are still governed by reason. It is reason committinig the accomplishment of an end to a trustworthy servant, that can do it better than she. Let that end -the
end indicated by the constitution of the appetites in their
relative positions - be accomplished, - no more, no less,and both reason and conscience are satisfied.
But besides the natural appetites, there are those termed
artificial, or, more properly, unnatural, as that for intoxicatimg drinks, for tobacco, and for opium. In all these the
principle is the same. An unnatural stimulus is given to
the nerves, followed by a corresponding depression, and an
uneasiness which causes a desire of repetition, and which
often becomes a craving so importunate as to overmaster
and control every other principle of action.
Between these artificial appetites and those that are natural there are four important differences.
90
ARTIFICIAL APPETITES.
The first is, that in the natural appetite the craving is an
original part of the constitution, created by God with reference to an end intended by him. In the artificial appetite, the craving is wholly superinduced by man, and with
reference to an end which God no more intended than he
did murder.
The second difference is, that the objects of the artificial
appetites are all violent poisons. They are incapable of
assimilation with the system. Except as medicines they
can contribute nothing to its health or well-being, and
taken in any considerable quantity they cause death.
The third difference is, that the pleasure connected with
the artificial appetites is purely and utterly selfish. It has
no relation to the ulterior good of the man himself, or any
other being. On the contrary, it lowers the tone of the
system and the capacity for good; whereas the pleasure
connected with the natural appetites has relation to the
vigor which wields the axe and guides the plough, and
even to the highest intellectual exertion.
The fourth difference is, that the artificial appetites have
a tendency to increase. As the stimulus is continued, the
quantity necessary to produce the desired effect becomes
greater. It is this insidious tendency, this "facilis descensus averni," that has brought many gifted men to the
verge of destruction before they were aware of it, and has
prevented their return. The natural appetites have no
such tendency.
Let no one, therefore, suppose that God has not given
as many appetites as are for his best good, or that he shall
be a gainer on the whole by attempting to reap where
nature did not sow.
The wretchedness there is in the world from the abuse
of the natural appetites, and from the expense and tyranny
91
LECTURES ON MORAL SCIENCE.
of the artificial ones, is so great that the purpose of God
with reference to this part of the constitution is worthy of
careful study.
We now pass to the Desires.
Of these the appetites are not only the condition, but
they foreshadow and symbolize them. The desires are to
the mind what the appetites are to the body.
Their negative characteristics are that they do not take
their rise from the body; that they are not periodical, and
that they do not cease after attaining a particular object.
Positively, they are cravings which have for their object the
well-being of the mind, as the appetites have for theirs the
well-being of the body. They act in the first place impulsively and specifically with reference to particular objects;
subsequently they are adopted by the reason, and through
the operation of that and the generalizing faculty, their
objects come to be designated by general terms, as knowledge and power.
What the original desires are, and how many, philosophers have not been agreed. This we may ascertain as
we may what the appetites are. The ultimate appeal must
be to consciousness; but if we can determine beforehand
or by observation what is requisite for the well-being of
the body, we can tell what the appetites will be. So with
the desires. If we can ascertain what is needed for the
well-being of the mind, we may know what they are.
Towards those things we may be sure there will be instinctive tendencies or impulses which reason is to accept,
direct, and limit, but which will not wait for the discovery
by her of their necessities before they act.
The desires, like the appetites, imply appropriation, a
gathering in, a use and assimilation of materials by ourselves. They are related to the affections, and are for the
92
THE DESIRES.
affections which are above them, and which imply bestowment, and giving out. As the appropriations by the appetites were not intended to be selfish or for their own sake,
but for the giving forth of every form of physical and
mental activity, so the appropriations by the desires were
intended to furnish the material and groundwork for the
activity of the affections and the will.
What, then, would be needed for the perfection of the
mind itself, and that man might act most effectively through
his affections for the good of others? He would need, -
1st. HIis own continued and secure existence. He would
need property, that is, the possession of those things by
which life may be sustained. He would need it both as
a provision for himself, and as a condition of generosity to
others. HIe would then need knowledge for his guidance;
he would need power to reach thie ends suggested by a
regard for his own good and the suggestions of the affections for others; and he would need the good-will and
esteem of others that he might cooperate with them, and
they with him, and stand in such a relation to them as to
be able to do them good. These he would need; they
would be indispensable to his completeness in himself, and
in his relations to others; and for each of these he has a
iiatural and original desire.
The desires, then, which we shall consider, are, 1. The Desire of Continued Existence.
2. Of Knowledge.
3. Of Property.
4. Of Power.
5. Of Affection, Good-will, Esteem.
Besides these, it has been said of late, and almost universally, that we have the desire of happiness, and the
desire of society.
93
LECTURES ON MORAL SCIENCE.
That the desire of happiness cannot be placed on the
same footing with the other desires, is plain,'
1st. Because happiness is the result of the normal activity of each of the faculties. We know it only as such.
But a desire, whose office it should be to receive the product of all the other faculties, would differ much from a
simple desire that produces happiness. In other cases the
desire is for a specific thing, and when that is met happiness is the result; but if we suppose an original desire of
happiness, there can be no happiness back of the happiness
desired, to be its result, and so its whole constitution must
be different from that of the other desires.
2d. It does not seem either simple or philosophical to
make a desire for knowledge, and a desire for the happiness resulting from that, each an original and simple desire. It would be more plausible to suppose, as some do,
that the desire for happiness is the only original desire,
and that the desire of knowledge, like that for books, is
wholly secondary. But this will not do, because, if we
had had no original desire for knowledge, we could never
have begun to seek it, and should have found no happiness
in its pursuit.
3d. In all other cases the desire goes directly to its own
object. It finds that, and happiness is the result. But no
man ever sought, or can seek, directly for happiness; he
must have something else as his direct object, and find
that indirectly.
4th. As each desire impels directly to its own end, and
knows of nothing else, it may, in a measure, be its own
guide; but, as happiness may result firom different and
often incompatible desires and faculties, there is far more
need of a higher power than any blind impulsion to guide
in its pursuit.
94
THE DESIRE OF HAPPINESS.
What, then, is the relation of this to the other desires?
To me it seems to be the same as that of consciousness to
the several specific faculties of cognition. Consciousness
is not a separate faculty, but accompanies and pervades all
the acts of each faculty. In the same way the desire of
happiness is not a separate and specific desire, but accompanies and pervades each act of such desire. As good is
the immediate product of the activity of our faculties, it
must be given in the original act of consciousness. Every
such act involves the conception, first, of being; second,
of activity, since consciousness is activity; and, third, if
the act be normal, of good as the result. But good thus
known must be desired, otherwise it could not be conceived of as good. In this way it is that a desire of good
enters into every specific form of desire, and that, as consciousness is the generic form of cognition, so the desire of
good or of happiness is the generic form of all the desires.
For the existence of a specific desire of society the
authority is high. That society is the natural sphere of
man there can be no doubt, and it is surprising that the
hypothesis of HIobbes, that the state of nature is a state of
warfare, should have been deemed worthy of a labored
refutation. "Man," it has been well said, "is born in society, and there he remains." The state of nature is a
state of society.
But, while it cannot be doubted that man was formed
for society, I yet esteem it rather a condition of his being
than the object of a specific desire. Hie has desires and
affections the exercise of which implies society, and it is,
as it seems to me, the direct exercise of these, and not
society itself apart from - this exercise, that he desires.
Take from him the desire of esteem, of power, of loving
and being beloved, all those specific desires, and affections,
95
LECTURES ON MORAL SCIENCE.
and sympathies, which are mentioned by the philosophers
separately, and which imply society for their exercise, and
the residuum that would be left of a desire of society,
as such, would be little or nothing. Observing a certain
effect, the combined effect of all our faculties, they seem
to have contrived a new faculty to account for it, extracting and compounding it from all the others. Ceitain it is that our delight in society arises chiefly from the
exercise of other desires and affections which there find
their sphere, and if any shall choose to say that there is,
besides the effect resulting from the combined influence of
these, an instinct or desire for society, I am content.
Though happiness and society are not inaugurated ana
guarded by a particular desire, yet the design of God in
regard to them is even more clearly and strongly indicated
than if they were. To me these seem to be, the one like
warmth, and the other like the atmosphere, pervasive and
enfolding conditions of our activity, and hence more intimately associated with it, and more fully cared for than
any single principle of action. They are like the axioms
in mathematics that are essential at every step in the reasonings, as compared with the definitions and hypotheses
on which particular demonstrations depend.
I shall close this lecture with some remarks on the first
of the desires mentioned,- that of Continued Existence.
This is often mentioned as the strongest of the desires.
We say, "as dear as life itself." Yet it yields to that of
reputation, and revenge, and sometimes gives way before
mere weariness and ennui. Nor is the fact that there are
so few suicides certain evidence of the power of this desire, since men often fear death greatly who desire life
feebly, or not at Ill.
It is the object of this desire to guard life in sudden
96
TRUE COURAGE.
emergencies, and to ensure for it our deliberate and
rational care; and our present business is to inquire how
far we should be governed by it.
This involves the question respecting a true courage,
since a man is to brave danger and to die when required
by that, and only then. Under no circumstances is a man
to be a coward.
It is the grandest characteristic of man that he can deliberately look death in the face, and accept it rather than the
alternative of spiritual degradation. On the earth there
has been no nobler spectacle than that of those to whom
this alternative has been presented, and who have chosen
to die, to die in torture and in the midst of reproach.
Required to renounce their integrity, or do violence to
their affections, they have chosen to become martyrs.
To die thus implies the conviction of an inner life far
higher and dearer than that of the body, which no weapon
can reach, and no flame scorch; of a liberty which no
manacles can restrain; and of a will which all the might
of nature cannot subdue; and the moment in which malice lifts its cry of seeming triumph over the destruction
of the body of one dying thus, is the moment of the
greatest possible triumph of fortitude and principle, and
of liberty in its highest form. That man is capable of
such persecution, is the greatest disgrace of our nature;
that he is capable of enduring and triumphing over it, is
its greatest honor. One such death, transcendent and
perfect, the world has witnessed; it can never witness
another. If we are called on to lay down our lives thus,
we are to do it as best we may. To do this is true courag,e; not to do it is cowardice. In doing this we become
martyrs; and no man has a right to do it, except as a
martyr to truth, to righteousness, to liberty, or to humanity.
9
97
LECTURES ON MORAL SCIENCE.
In imitation of this, but in striking contrast with it, is
that common-place exposure to danger and death which
comes from recklessness, and vanity, and a regard to the
opinion of others. There can be no nobleness in blindfolding the eyes, or in suffocating the natural emotions.
Rightly viewed, it is an awful thing to die. It becomes
us to acknowledge this; but if required to testify to any
great truth, or to sustain any great principle, it becomes us
to have such a conscience, and such a trust in God, that we
may die without fear, or even with welcome. This is true
courage, and anything else in the guise of this is either
stupidity, or cowardice and hypocrisy.
But the obligation to meet death with firmness, when
called to it by truth or by duty, does not rest solely upon
our individual interests and character; the interests of
mankind are involved. Abstract truths and general principles often lie dormant till they are awakened into life
by some powerful attestation. The attestation which the
death of a wise and good man gives to the value of the
principles for which he dies, has a voice that is startling to
humanity, and will arouse it if anything can. If the existing generation do not hear it, as through interest or
prejudice they may not, it will not be lost; it will be
heard in after times. It is for this reason that the blood
o-f the martyrs has been the seed of the church, and that
the names of IHampden and Sidney have been the watchwords of liberty wherever the English tongue has been
known. When such men die, death, in whatever form,
does not come to them as to common men, whispering of
terror or of hope for them alone, but -
" In its hollow tones are heard
The thanks of millions yet to be."
However strong, therefore, the desire of life may be, it
98
DESIRE OF CONTINUED EXISTENCE.
must yield when this is required by higher principles of
action, by the affections, and the conscience. Mankindc
justify and applaud him who dies for his kindred, his country, his race, or to sustain his integrity. They disregard
and despise him who dies, or exposes himself to death,
from a desire of applause, or from the fear of a corrupt
public opinion.
It only remains to notice the modes in which the intentions of God, as indicated by this part of our constitution,
are plainly set at nought. These are chiefly four.
The first is, by any vicious indulgence which shortens
life. The guilt and waste of life from this cause cannot be
measured.
The second is from war. We need not inquire here
whether men may expose their lives in war according to
the principles already stated. That they may not on lower
principles, is certain; and in the light of this truth, how
dreadfully have the purposes of God in regard to human
life been disregarded in war! So has it been in all wars
of ambition, of passion, and of mere interest. The fact
that mercenaries have been so readily found, who would
espouse any cause, expose themselves to any danger, and
do any amount of slaughter for the poor pittance of a soldier's pay, is among the saddest indications of the moral
state of the race.
A third mode in which the purpose of God, as indicated
by this desire, is set aside, is by suicide.
As this is a crime which cannot be punished, little can
be done to prevent it except to point out and remove its
causes. These are, -1st. Insanity. With this we have
nothing to do. 2d. The commission of crime and apprehended exposure and disgrace. 3d. Disappointment in the
attainment of any object which has been regarded as the
99
LECTURES ON MORAL SCIENCE.
chief good. 4th. Infidelity when carried to the denial of
a hereafter or of human accountability. Not that infidelity has a direct tendency to induce suicide, but that,
when men are tempted to it, it removes all obstacles. A
thoroughgoing and unflinching infidel would feel himself
at perfect liberty to choose nonentity rather than life if he
should prefer it. lIence the levity with which this crime
is spoken of by infidels, as Hume, who said that it was
but the turning a little blood out of one channel into
another. It is only by the removal of the causes now
mentioned that we may expect that the frequency of this
crime will be diminished.
A fourth mode in which life is wantonly shortened is by
duelling.
In this we have a striking instance of the power of
custom after the opinion in which the particular custom
originated is entirely changed. Originally regarded as a
species of judicial trial in which there was an appeal to
God, a refusal to fight came in time to be considered a
confession not simply of cowardice, but of cowardice on
account of guilt. Then it was that the tyranny of custom
and of public opinion commenced; and now, though the
idea of an appeal to God, or of any adjudication according
to merit, is utterly exploded, though the laws are against
it, and it is known to be morally wrong, though the force
of public opinion is in some regions entirely removed, and
everywhere very much lightened, yet the custom still retains its hold, and the law of God is made void by the
"traditions" of men in high places. This, too, is done
when all the circumstances which once gave the combat
eclat and dignity are entirely reversed. It was once sanctioned by law, and witnessed by multitudes who applauded
the knightly bearing of the combatants. Now, those who
100
DUELLING -MORAL COWARDICE.
fight shrink away to some place where the law may be
evaded, the combat is witnessed only by the seconds and
the surgeon, and there is no display of manly vigor, or of
any other skill than that of a highwayman. The parties
simply take pistols and shoot at each other. It was once
an evidence of courage, and compatible with a sense of
duty; now, whatever may be said of mere animal courage,
it shows a pitiable want of moral courage, and is opposed
to all the dictates of morality, of humanity, and of reli gion. Though founded in mistaken notions, it yet had, at
its commencement, something noble about it, but like the
Scylla of Virgil, whose head was human, it tapers off, as it
comes down to us, into hideous and unmitigated deform' ity. In its present position, it is difficult to say whether
this custom is more wicked or ridiculous.
9*
101
LECTURE V.
DESIRE OF PROPERTY. - AVARICE. - DESIRE OF NOWLEDGE. - DESIRE OF
POWER. - INFLUENCE. - EMULATION. - DESIRE OF ESTEEM. - DESIRE OF
GLORY.
AFTER the desire of life, which we have already considered, that of property was mentioned.
As life is the condition of all the desires, so also is the
possession of that which is necessary to sustain life. In
common with the others, this desire has its root in the
tendency of all life to appropriate to itself whatever is
necessary to its own perfection and manifestation. So it
is with the appetites as they are related to the perfection
and power of the body. There is a point where they are
identical, and whence they branch off in search of different objects necessary for such perfection and power, and
so become different specific appetites. So, also, it is with
the desires. There is a point where they, too, seem identical in their relation to the perfection and manifestation
of mind, and whence they branch off in the directions
mentioned as constituting the several specific desires. If,
therefore, the ownership of something, possession, property, be essential to such perfection and manifestation,
then this general tendency will be in that direction, and
will become a specific desire.
But ownership, or property, is thus necessary. It is
through this that we have security for ourselves, and a
chief means of manifesting our individuality to others.
102
0
THE DESIRE OF PROPERTY.
What is not our own we have no right to use. We have
a right to use the fruit that grows wild only because, when
we pluck it, it becomes ours. And, as this sense of property is the condition of our using anything for ourselves,
so is it for our giving anything to others.
We may, it is true, conceive of a state in which the
whole enjoyment of man, and perhaps an adequate one,
should arise from what could not, or need not be appropriated, as the air and the sunlight; but, in his present
state, if he had no material thing which he could use as
his own, and none which he could give to others, he not
only could have no security, but would lack scope for the
activity of some of those essential faculties by which he is
made in the image of God. If God had no ownership, he
would not be God, and if man had none, involving dominion, he would not be in his image.
That the desire of property in the sense and to the
extent above indicated is a natural desire, we can scarcely
doubt, if, in addition to the considerations just adduced,
we notice how early and distinctly it is manifested by children; how it stimulates industry; and how essential property is to the very existence of society. Doubtless, the
natural desires often interpenetrate, support, and modify
each other, but there seems to be no more reason for referring this desire, as some have done, to that of power, than
for referring the desire of knowledge in the same way,
since knowledge has often been said to be power. Holding
such relations as property does, we might expect that God
would indicate his will by giving a specific desire, and that
he would make that desire, as he has all the others, the
basis of a right. If God has given us a desire for property, then, within limits to be fixed by other considerations, we have a right to property, and when we look at
103
LECTURES ON MORAL SCIENCE.
the extent and validity of the right of property, we can
hardly suppose it to be founded on anything but a natural
desire.
This desire, then, being, in the true and original sense of
that word, natural, cannot be wrong. Nor is it too strong
in itself, for there is not too much honest industry or selfdenying frugality. The doctrine holds here, that has already been stated in regard to those principles of action
which relate to the material interests of the individual and
of society, that the stronger they are, provided they be kept
properly subordinated, the richer and better substratum of
individual character and of society do they form. Those
who have done the most for our public institutions, and
done it most nobly, have been men with a strong desire of
property, who knew the worth of what they gave; generally men who had accumulated it by their own industry,
but who gave, nevertheless, cheerfully and gladly, in view
of great interests to be promoted, and of the subordinate
place which this desire holds as the purveyor of God, and
the appointed servant of principles higher than itself. If
an alabaster-box of precious ointment is to be opened, the
perfume of which is to fill society, the box must first be
filled. Only as we recognize the legitimacy of this principle can giving have its true merit and dignity, or indeed
any merit or dignity at all. As men now are, it is far better that they should be employed in accumulating property honestly, to be spent reasonably, if not nobly, than
that there should be encouraged any sentimentalism about
the worthlessness of property, or any tendency to a merely
contemplative and quietistic life, which has so often been
either the result or the cause of inefficiency and idleness.
But while the legitimacy of this desire is not to be ques
104
DESIRE OF PROPERTY.
tioned, it is not to be forgotten that it is specially liable to
excess and perversion.
The appetites have a material limit, but, like all the desires, this has none, and, unchecked, it grows, and becomes
insatiate by its own activity. It is like an elastic receiver
which could not be stretched beyond its capacity, but
which would grasp the more tightly its contents the fuller
it should be made. To the strength of natural desire
there is added the power of habit; and then, in our state
of society especially, there is everything to foster it.
With no law of entail, with a form of government that
stimulates every faculty, with unprecedented openings for
enterprise from the newness of the country, with no order
of nobility, and, with the exception of high talent and
transient office, with nothing but wealth to give position
and distinction, it is not strange that it should be sought
with peculiar eagerness and unscrupulousness. More than
any other it is the national passion, and, what with dishonest and injurious modes in the getting, and folly and
luxury in the using, there is danger through it of national
ruin. It is not merely on the protection of the right of
property, essential as that is, that the material prosperity
of a nation depends, but also on the prevalent modes of
getting and using it. Gambling, lotteries, theft, fraud, are
modes of gaining wealth, but are mere depredations on
society; pandering to hurtful and vicious appetites is still
worse, and when these are prevalent, implying as they do
modes of spending money corresponding with the modes
of getting, there can be no prosperity.
The perversions of this desire appear in covetousness
and avarice. These have in fact the same elements; but
covetousness, even to unscrupulousness, in getting property, is not incompatible with profusion in spending it;
105
LECTURES ON MORAL SCIENCE.
while avarice refers more particularly to the grasp with
which it is held. This grasp may be so strong as not only
not to be relaxed at the call of public spirit and natural
affection, but even for the supply of the most pressing personal wants.
It is here that we find, and are called upon to account
for, that strange phenomenon in our nature, - a miser. A
miser is one in whom this desire is so strong as to defeat
all the ends for which it was given, one who suffers the
very wretchedness which the desire was given to prevent,
through an excess of the desire given to prevent it.
As it is money that is especially sought by the miser, it
has been usual to say that as that is the representative of
value, and stands for everything which it can command, we
transfer, through the association of ideas, the regard we
have for those things to that which represents and can command them, and so come to attach a high intrinsic value to
that which has little value in itself, and none at all so long
as it is hoarded. That something of this occurs almost
universally, cannot be doubted, and if we combine it with
unusual outward temptations, or with peculiar constitutional tendencies, or both, it may be sufficient to account
for many cases of miserliness. Doubtless there are those
to whom this is naturally a besetting sin. But there are
cases for which it does not account; especially those in
which persons who have been prodigals in youth have subsequently become misers. This has often been the case.
It was so with the noted miser mentioned by Foster in his
Essay on Decision of Character. But of all men we should
suppose a prodigal would be the last to associate money
with value. Brown, therefore, founds avarice, not so
much on feelings of pleasure at seeing constant additions
to a heap that is never to be used, as on the permanence of
106
THEORY OF AVARICE.
money compared with the transient pleasures of the prodigal, and on feelings of regret at having spent that which
can never return. If a man purchase a house, though his
money be gone, yet the house remains, and being constantly useful to him, he looks back upon the parting with
his money without regret. But if he had expended the
same sum for a palace of ice, though he might be pleased
for a time with its glitter, yet, when it had melted away,
he could not fail to reflect how much that was valuable he
might purchase with his money if he then had it, and look
back upon his parting with it with regret.
Let, then, a young man spend his money foolishly till
he becomes embarrassed, or perhaps in utter want; let
him be stung at the same time by what is, or what he conceives to be ingratitude, and every instance of such expenditure will haunt him, and a permanent and deep feelingof regret will be the consequence. If he again acquire
money, he will regard it not so much as the representative
of any particular value, as a guard against the perplexity
and trouble into which he had previously fallen. As he
formerly reflected afterwards how many things he might
have purchased, so now his money seems to him, not the
representative of the value of that particular thing which
he may wish to purchase, but of all those things collectively which might be obtained by it. As it was from
parting with his money that his regret formerly arose, so
now, when he would part with any, whether the sum be
great or small, and quite as much if small if it was by
small sums that he lost his money, the same feeling presents itself and debars him, till at length penurious and
miserly habits are formed.
This theory I deem correct, and bring it forward for the
practical moral consequencs which it involves. It is often
107
LECTURES ON MORAL SCIENCE.
thought an indication of spirit in young men to have a certain profusion and recklessness in their way of spending
money. They think it essential to their position to spend
upon trifles of fashion, and the demands of what is called
good-fellowship, but what is too often fellowship in folly
and vice, sums which neither they nor their friends can
well afford. If, then, instead of being considered a mark
of spirit, this profusion were regarded by the young man
and his friends, as it truly is, as a mark of want of judgment and of genuine independence, and if in the prodigality of to-day they could behold the parsimony of future
years, much evil would be averted.
In our cities and public institutions there are many
young men who depend on a hard-working father, or a
poor and widowed mother, or on self-denying sisters, who
are liable to be drawn into associations with those whose
means of expense are above their own, to incur obligations
of what they call honor, and to engulf, if not in vice, yet
in what is purely conventional and useless, the scanty earnings of their home. It is pitiable to see those who do
thus, greedy of money whenever they can get it, evading
small bills, and those of poor people; disappointing, alienating, perhaps ruining those who love them; losing their
own self-respect, and incurring the contempt of those who
care little or nothing for them. From such the public has
nothing to hope.- But from one who will deny himself,
and rely for his position upon industry, integrity, and
transparency of character, and who can respect himself in
honest poverty, and look down upon meanness anywhere,
if he shall succeed, the public may expect much. He will
have an open hand for somebody.
In general, if we have been accustomed from our youth
to spend money so that we have not regretted its loss, if
108
THE DESIRE OF KNOWLEDGE.
we have given it for the necessaries and the conveniences
of life, and especially for the gratification of the benevolent affections, we may expect to continue to part with
money,- that is, if we have it,- if not nobly, yet usefully, and without regret. But if we have spent our
money aimlessly, or with that mixture of meanness and
profusion which those often exhibit who spend money only
for selfish pleasures, we must beware lest the reckless
expenditure of twenty become the avarice of sixty; lest
the young man, flattered and praised by sycophants for his
generosity, become in age a niggard and contemptible
miser.
From the desire of property we pass to that of knowl
edge.
By the first we appropriate to ourselves whatever may
be useful to us that is material; by the second, so far as
that is possible, whatever may be useful that pertains to
the spiritual world.
That this is a natural desire need not be proved, because it is not disputed. This was known to Solomon.
"Through desire," says he, "a man having separated himself, seeketh and intermeddleth with all wisdom." Like
him we give our hearts "to seek and search out by wisdom concerning all things that are done under heaven."
It may be "a sore travail," but "this," in giving this desire, "hath God given to the sons of men to be exercised
therewith." The desire has for its object the only element
in which man can walk without stumbling. It is as the
light by which we see, and so is indispensable to the intelligent exercise of any of the faculties.
It is not to be supposed, however, that all knowledge is
gained under the stimulus of this desire. The desire is
10
109
LECTURES ON MORAL SCIENCE.
found existing in the light of consciousness and of the
primitive ideas and truths of reason. These, which have
been said to be
"The light of all our seeing,"
are essentially the same in all. They are involved in the
exercise of all our faculties, while this desire of knowledge,
or the principle of curiosity as it has been called, may
exist in different degrees, and with reference to different
objects.
So far as the desire of knowledge is impulsive and involuntary it has no moral character. In this respect it is on
the same footing with all the impulsive powers. They
respect objects which are indifferent in themselves, that
may be used for either good or evil, and moral character is
manifested as we reject or adopt and control these impulsions. An angel and a fiend may have equal knowledge.
Their character is shown by its use.
Of this desire the direct and proper stimulus is knowledge itself, and for itself. To the mind that can feel it
there is in knowledge a power to charm as there is in
music. It is a high attribute of man through which he
can find in the works of God, and in the relations which
he has established, an excellence so attractive as to be in
itself a sufficient motive to their contemplation and study.
In this is the root of the true enthusiasm for science. It
is among those who have this that we find the mathematicians, who, like Archimedes, can spend days and nights
in the contemplation of abstract theorems; the sages, who,
like Socrates, can remain absorbed in thought four-andtwenty hours without changing their position; and without much of this no man can be expected to distinguish
himself greatly in the walks of science.
110
THE DESIRE OF KNOWLEDGE.
But besides this primary motive, the desire of knowledge finds a natural and legitimate support in the esteem
in which those are generally held who are distinguished by
their attainments; in the direct and obvious utility of
many branches of knowledge, and, from the wonderful and
often unsuspected connection of its different branches, in
the incidental and possible utility of all knowledge.
But even with such support, the desire of knowledge
has often too little relative strength in the contest with
indolence. In order to induce study, the best of men
have therefo)re thought it necessary to admit and to sanction in our public institutions the far inferior and sometimes pernicious motive of emulation, but they have done
it reluctantly, and only as polygamy was allowed to the
Israelites, "because of the hardness of their hearts." It is
to be hoped that the time may come when the adjustment
of forces shall be different, and there shall be found in
knowledge and in its necessary and legitimate results sufficient motive for its pursuit.
Like the appetites, the desire for knowledge may become
artificial, and take directions that are capricious. It may
also be in excess. It is always relatively so when the
acquisition of knowledge has no respect to the attainment
of mental power, and the use to be made of it. Knowledge is the food of the mind. And as food may overload
and enfeeble the body, and is to be received only as there
is a capacity of digestion and assimilation, and with ultimate reference to action, so knowledge may overload and
enfeeble the mind, and should be received only as it can
be reflected on and arranged, and so incorporaterimto our
mental being as to give us power for action. Here, as
elsewhere, the receiving is to have reference to a giving,
but not wholly. If the thing received were not valuable
i1l
LECTURES ON MORAL SCIENCE.
in itself, there would be neither worth in the gift nor
merit in the giving.
We now proceed to the desire of power. The idea of
power is inseparable from that of will. The very act of
willing, or, as Hamilton calls it, conation, gives the conception; but this is fully realized only in the passing of the
conation into its results. Personal power is in the idea of
a will; but the idea of power is implied, and would be
given also in the spontaneous exercises of any of the faculties. A faculty and a power are the same thing. It
may even be said that in all receptivity there is power.
There is the power of receiving; but in the sense now
contemplated this would not be a power. In all power
exerted there is an origination of activity.
The idea of power, then, enters into our very conception of ourselves. We cannot exist except as powers.
The consciousness of being, and of power, can hardly be
said to be two things. We can neither know nor rejoice
in our being nor its enlargement except through a consciousness of power, and of the enlargement of power.
Doubtless there is a high pleasure as we make experiments
upon our faculties corporeal and mental, and ascertain the
effects we can originate through them, and the more striking the effects the greater the pleasure; but in that we are
merely finding ourselves out, and the desire of power no
more respects, as has commonly been supposed, that power
which enters into the conception of ourselves, than the
desire of knowledge includes the light of consciousness
and the intuitions of reason. The very desire of power is
itself a power, and it is absurd to say that a desire desires
itself. In examining man we must take him as possessed
of all that makes him man. We find him to be a power,
112
DESIRE OF POWER AND OF LIBERTY.
and also to possess the desire of power; but, as the desire
of knowledge implies a primitive knowledge which the
desire does not respect, so the desire of power implies a
primitive power which that desire does not respect. We
must either adopt this mode of viewing the subject, or
resolve all our desires into that of power, since there is in
all of them an exertion of power and an enjoyment proportioned to the power exerted.
But all this is very different from that control over nature and men which we may gain by our own skill and
exertion, which may be put forth in different directions, or
not at all, and the desire of which may exist in different
degrees. It is this, and chiefly the desire of controlling
our fellow-creatures, that we mean by the desire of power.
Having made this distinction, it may be well to indicate,
at this point, the difference between the desire of power
and that of liberty, as the latter is often made a part of
the former. Liberty has no particular connection with the
desire of power as just defined, but has respect to the putting forth, within their legitimate sphere, of any of those
faculties by which we are men. It is the condition of the
manifestation of our being in any direction we may choose;
but I did not class it with the specific desires for the same
reason that I omitted the desire of society and of happiness.
A free bird does not desire freedom. It was hatched firee.
Freedom is the general condition of all its activity. So
men are born free, and God has given no natural desire to
meet a condition of things induced by wrong. It is, therefore, no specific desire, but the whole nature, that rebels
against unjust restraint; and freedom can be crushed out
only by the degradation of the whole man.
Hence liberty, society, and happiness, the first two being
general conditions of our activity, and the last a general
10*
113
LECTURES ON MORAL SCIENCE.
result of it, are more intimate to us, more essential and
sacred, than the object of any specific desire.
But to return. The power of man is from his will, and
the extent of it he learns wholly by experience. If the
movement of mountains had followed his volition from
the first as uniformly as the movement of his limbs, it
would have seemed to him, and would have been, no more
strange. But experience shows him that his direct power
extends only to the voluntary muscles of the body and to
the voluntary faculties of the mind, and that even here his
power is not absolute. Probably no man ever gained the
full control either over his muscles or over the faculties of
his mind. To give such control is one great object of education. In this is discipline. IHere is the first sphere of
power, the only one that is direct. Here lies the greatness
of him who ruleth his own spirit.
But between this power, which, though direct, is so narrow in its range, and that indirect power which man may
exert over the elements and over nations, the contrast is
marvellous. It is this latter power that men chiefly seek,
and all mechanism, all practical sciences, all forms of government, are but means for its exercise. They are means,
more or less facile, for connecting the will of man with
remote results; and nothing more indicates the superiority
of man's nature than the extent to which this may be
done. An animal can do nothing at a distance from itself
in space, and nothing worthy of mention, except in the
present, in time; but the will of a single man may find
expression in a few words that shall set in motion armies
and navies, and the echo of what was at first but a few
feeble vibrations of the atmosphere shall come back from
distant continents in the roar of cannon and the groans of
the dying. The thought and feeling of one man may find
114
THE DESIRE OF POWER.
expression in words that shall be repeated through all
time, and work like leaven in transforming society.
The control of man over nature can never be arbitrary.
"Nature is conquered only by obeying her laws." But
while nature cannot be broken down by force, the will of
man may be. Hience, in governing his fellows, instead of
making, as he should, the method he is compelled to follow in nature his model, and governing men in accordance
with the laws of their rational and responsible nature, man
has employed arbitrary power. It is in the tendency to
this that the danger from this desire is found. To a corrupted will the taste of it is like that of blood to the tiger.
Under its influence man sets himself up as independent
of authority, rejects moral restraint, and in passing to his
selfish ends disregards the rights and the miseries of men.
The larger part of history is but a record of the deeds of
men under the influence of this desire thus perverted.
But whatever the perversions and abuses of this desire
may be, there can be no more doubt of its legitimacy than
of that of knowledge, since the great use of knowledge is
to be a condition for the right exercise of power. If the
results of its perversion are terrific, it only shows the uses
to which it may be put when rightly directed. The element that rages in the conflagration is the same that
enables man to mould to his will the most refractory substances in nature, and which may be made so much the
more energetic in its usefulness, as, when uncontrolled, it
had been destructive and awful. It is the same atmosphere that, in its condensed energies, forms the tornado,
that wafts the ship, and kisses the leaf of the violet.
Every creature of God is good; but it is to be used not
only "with thanksgiving," but in accordance with his
laws.
115
LECTURES ON MORAL SCIENCE.
It is this arbitrary power that kings, and especially
tyrants, have sought and possessed. In the early stages
of society it was natural that such a power should be
attained, and, once attained, that it should seek to perpetu ate itself. But men have found the trust too great. They
are, therefore, seeking to divide the power, and, by putting
it into the hands of the people themselves, to bring inter est in to the aid of principle. It is creditable to man
that he can maintain a republic, but it would be more
so if monarchy could be well administered. If democracy
trusts the people to a certain extent, it yet proceeds upon
a distrust of man. By adopting as its maxim that "eternal
vigilance is the price of liberty," it declares that man has
no moderation in the desire, and is not to be trusted in
the use of power.
But besides arbitrary power backed by force, there is
another which we call influence, not less effectual, and
often not less extensive, which is exercised, not by coercion, but in compatibility with the laws of mind and the
freedom of others. This is the power of the wise, of the
eloquent, of the good man; and as it always implies the
possession of qualities respectable in themselves, and generally beneficial, it is to be sought by every honorable
means.
Power and influence are not incompatible, but, as contrasted, they differ in several respects. Power interferes
with freedom; influence does not. Power stands above
those whom it controls, and issues its commands; influence elicits and directs the individual energies of those
upon whom it bears, and thus enlarges the sphere of their
agency. Power keeps itself aloof as an object of fear and
admiration; influence mingles in with the agencies which
itself has set in motion, and is often so lost in them as to
116
EMULATION.
be forgotten, as the kindling spark is forgotten when the
flames begin to spread. Power, especially if it be hereditary, depends upon accident; influence upon personal
qualities. Power is maintained by pageantry, by chicanery,
by brute force; influence by the cultivation of those commanding qualities from which it first arose.
While, therefore, we reject, as the object of desire, all
arbitrary power, we cannot too earnestly desire those
means of influence by which we may lead others freely to
their own good.
Emulation, or the desire of superiority, is classed by
Stewart and others among the original desires. By others
it is regarded as a modification of the desire of power.
So I regard it. At least I hardly know where else to
place it, though the desire of esteem often seems to be
involved in it, quite as much as that of power. If the
contests in which emulation is excited were not public, and
the results were never to be known, probably the emulation would be but slight. My reason for not classing it
with the original desires will be found in the principle
already stated. I do not see that it would be necessary to
the perfection of the mind.
Of this as a principle of action much has been said, and
moralists are not agreed respecting it. This may be, in
part, from some ambiguity in the term. There can be no
emulation unless a man pursues an object in common with
others. Here other principles are brought in, and we
need to discriminate.
There is in many animals an instinctive feeling that produces in them the effects of emulation. It may be seen in
two horses drawing together, or attempting to pass each
other. This feeling has in it nothing malignant. It is
probably a modification of their social instincts.
117
LECTURES ON MORAL SCIENCE.
In man there may be something of the same instinct,
joined with the higher influence of sympathy. Of sympathy the influence is so great that Adam Smith made it the
foundation of his moral system. If we see others laugh,
we are disposed to laugh also; if they are in grief, our
feelings and countenance conform in some degree to their
emotions; and whatever feeling may be vividly expressed,
if it does not shock our sense of propriety, we have a tendency to enter into and sympathize with. This is natural
and right. If, now, in a class of young men studying together, and doing as little as possible, we suppose that one
of them should wake up to a love of knowledge, and to a
sense of his responsibility, and enter independently upon
a course of work, it would be strange, since we sympathize
with almost every other feeling, if something of his spirit
should not be transferred to others. So far from being
wrong in them to feel it, it would imply a baseness if they
did not, and if this feeling should pervade the class, it
would be a blessing to all. It would be simply a manifestation of our social nature in one of its higher and better
forms. That there is in it nothing of malignity or personal feeling is clear, because the same feeling may be
excited by reading the lives of those who are dead. What
was it that brought tears into the eyes of Julius Cesar,
when, at the age of thirty-two, he saw the picture of Alexander the Great? What is it that causes the bosom of the
young missionary to burn when he reads the lives of Brainerd and of Martyn? And if we may be thus stimulated
by those who have gone before uis, how much more by
those who walk with us. It is in this effect and propriety
of sympathy that we find not only the benefit of social
study and work, but the obligation of setting a good example. If any deny the propriety of being stimulated, not
118
EMULATION.
merely in view of the thing to be done, but also in view
of what others have done, they destroy the obligation to
set a good example. This principle is recognized in the
Bible "Consider," says the apostle, not simply the excel lence of the end, but "one another, to provoke - yes provoke -unto love and good works." "I speak not this,"
says he, "by commandment, but by occasion of the forwardness of others." He says, too, by way of commendation, and as what he rejoiced in, "And your zeal hath
provoked very many " not to do more than others, but
what they could.
Thus, when we pursue an object in common with others, our motives are mixed. We have some love of the
thing itself, we have some sympathy, some desire of the
esteem connected with distinguished success, and we may
also have a desire of superiority for its own sake. It is
this last only that is properly emulation. So it is defined
by Butler, and Reid, and Stewart, and Whewell; but in
supposing this to be an original part of our nature, and in
their discussions upon it, I cannot believe that they wholly
separated it from the elements above mentioned.
That this love of superiority, taken by itself, is either
a natural or a justifiable principle, I cannot suppose. It
does not contemplate our doing what we can, which is all
that is required of us, but more than another, and involves
our unhappiness if we do not. It is nowhere commanded
in the Bible that we should be above others. To desire
to be above him simply for the love of it, is incompatible
with loving our neighbor as ourselves. It is a pleasure
gained at his expense; but there is no legitimate pleasure
that is necessarily at the expense of another. God has
not so constituted his creatures. It is closely, though perhaps not necessarily, associated with pride on the one hand
119
LECTURES ON MORAL SCIENCE.
and envy on the other. It cannot blend with that love
which is the fulfilling of the law. To suppose one person
to be endeavoring to love God more than another is preposterous.
It is supposed by some that emulation is forbidden in the
Scriptures, because "emulations" are classed by the apostle Paul with " wrath, strife, envyings, murder," etc. But
in the Scriptures language is employed with the same latitude as in common life. The term is found in them but
twice, and in the other instance is used by the same apostle as that which he was desirous of producing. "If," says
he, "by any means I might provoke to'emulation' them
which are my flesh, and might save some of them." There
is, therefore, an emulation to be comnmended as well as one
to be condemned; and, doubtless, men often dispute on
this subject, who, if they would be careful to understand
each other, would find themselves perfectly agreed.
We have now considered, in its various forms, the desire
of power. The vanity of those pursuits to which men are
impelled by it, when in excess, is a common topic with
moralists. Doubtless, the objects of it are less valuable
when attained than they appear in the distance. The elevation is apparently smooth and inviting, but the way to it
is hazardous, and when reached it is often found barren
and comfortless. That those who enter upon this pursuit
should be deceived is almost a necessity. By men who are
in power, and have wealth, while they seem to have everything at command, their care, their weakness, their misery,
are carefully concealed. They often spend more thought
and labor to appear to be happy than to be so. Than our
judgments respecting the happiness of others nothing can
be more uncertain. The evils that we do not see we readily suppose not to exist, and often envy those who are far
120
VANITY OF WEALTH AND POWER.
more wretched than ourselves. The impression of pain is
much more vivid than that of pleasure, and a man apparently happy may have his life embittered in a thousand
ways which we do not suspect.
But, laying aside the evils common to all men, power
and wealth have cares and troubles peculiar to themselves.
"The needy traveller, serene and gay,
Walks the wild heath and sings his toil away;
Does envy bid thee crush the upbraiding joy?
Increase his riches, and his peace destroy.
Now fears in dire vicissitude invade,
The rustling brake alarms, and quivering shade,
Nor light nor darkness brings his pain relief, -
One shows the plunder, and one hides the thief."
"For heaven's sake let us sit upon the ground
And tell sad stories of the death of kings;
How some have been deposed, some slain in war,
Some poisoned by their wives, some sleeping killed -
All murdered, -for within the hollow crown
That rounds the mortal temples of a king
Keeps Death his court, and there the Antic sits,
Scoffing his state, and grinning at his pomp,
Allowing him a breath, a little scene
To monarchize, be feared, and kill with looks,
Infusing him with self and vain conceit,
As if this flesh which walls about our life
Were brass impregnable; and honored thus,
Comes at the last, and with a little pin
Bores through his castle-wall- and - farewell king.
Cover your heads, and mock not flesh and blood
With solemn reverence."
If these and similar evils of wealth and power are more
than compensated by peculiar advantages, the balance in
their favor is but slight. What is most to be desired and
11
1 1
LECTURES ON MORAL SCIENCE.
most to be dreaded in life is common to all men. The
light of heaven, the air, the earth, the heritage of the
senses, the play of the affections, the treasures of a good
conscience, may be possessed by all. From the loss of
friends, the encroachments of disease, the disorder of the
passions, the forebodings caused by sin of an awful future,
and from death, none are exempt. Where, then, there is
so much in common, the difference of enjoyment that mere
wealth or power can give is so small that if it must cost
much struggle it will generally be found that the "play is
not worth the candle," that we have sacrificed ease and
independence to imaginary advantages.
It only remains to speak of the desire of esteem.
For this the other desires are, in a measure, the condition, since esteem is most fully reached through the use
we make of property, knowledge, and power. It has reference not only to our own happiness, but to our cobperation with others, and is an indispensable condition of the
social results intended by God. It is less stirring than the
desire of power, and often requires us to forbear action as
well as to act. With the desire of arbitrary power it is
incompatible. He who would employ the means requisite
to gain that, and would use it when gained, must forfeit
esteem. Napoleon is reported to have said of his brother
Joseph that he was too good a man to be a great man.
That this is a natural desire, is not now questioned. It
appears in children before they are able to speak, and with
many is stronger than any other, even than that of life
itself or of a good conscience. Men will sacrifice life for
the good opinion of others, and will lie that they may not
be thought liars. Its opposite, scorn, contempt, ridicule,
are among the things we most dread, and it requires the
sternest principle and the greatest independence of judg
122
O
THE DESIRE OP ESTEEM.
ment to stand before them. To do this is a higher form of
heroism than to stand before the cannon's mouth. Few
will not remember the impressions from first reading Milton's description of the faithful angel with whom
" Nor number nor example wrought
To swerve from truth, or change his constant mind,
Though single. From amidst them forth he passed
Long way through hostile scort, which he sustained
Superior, nor of violence feared aught;
And with retorted scorn his back he turned
On those proud towers to swift destruction doomed."
From the legitimate influence of this desire the benefits
are equal to its strength. The danger also is in the same
proportion. This arises from the want of coincidence between the desire and the conscience in others and in ourselves, and will be in two directions.
In the first place, we may be desirous of doing right, but
be tempted to violate our conscience in order to please
others. This we are never to do, either by evasion or
compliance. Those who do this are a kind of inverted
hypocrites, seeming worse than they are. In matters of
indifference we are to be ready to comply with the inclinations, and even the prejudices of others, but if we violate
our conscience we not only incur guilt, but are generally
despised by the very persons whose good opinion we seek.
Besides, it is not to the good opinion of men only that
we should have respect. Many things that are highly esteemed among men are abomination in the sight of God.
This often causes a fearful conflict, but there must be no
faltering.
In the second place, we may suppose others to be pleased
with good qualities, and be tempted to make a pretence of
those we do not possess, thus violating our conscience by
123
LECTURES ON MORAL SCIENCE.
acting a lie. This takes two forms. If admiration be
sought, it will be affectation; if confidence and friendship,
hypocrisy. In both we act a lie, but the one is a ridiculous lie chiefly hurtful to ourselves, while the other is a lie
of the darkest hue. Affectation and hypocrisy! To how
much light satire and spleen, to how much deep distrust
and dark misanthropy, have they given rise! How have
they given to human life, i-which such momentous interests are involved, the appearance of a masquerade and a
farce!
Has any one, then, principle? Let him abide by it.
Would any one seem to be anything? Let him be that
thing. This is the freest and safest way, and quite as
easy as to preserve a state of forced and dangerous concealment. Regarding these two cautions, we need not
fear being too much influenced by a regard to the good
opinion of those around us.
The esteem spoken of hitherto is that of those whom
we know, and with whom we have intercourse. But we
also desire the good opinion of those who are remote from
us in space and in time, whom we never expect to see or
to have intercourse with. We desire fame, and, what is
the highest form of it, glory.
By some this form of the desire of esteem has been
ranked as a separate desire, but without reason. By others
it has been greatly ridiculed, also without reason, since it
is a natural form of the desire, and one justified by the
Scriptures. "The righteous," they say, "shall be had in
everlasting remembrance;" and Christians are those who
seek for "glory and honor," as well as for "immortality."
Of glory as it is commonly conceived, Cousin has given
the best account I have seen. That I propose to give in
substance, and then make some remarks upon it.
0
124
GLORY AND REPUTATION.
And first, we are to separate glory from notoriety. The
passions and feelings of one man are common to all, and
mankind are always aroused by any vivid and startling
exhibitions of their common nature in any of its elements
or forms. If this exhibition be of the darker and fiendlike passions, they will utter a cry of execration which is
at once notoriety and infamy.
We must also distinguish glory from reputation. This
implies something praiseworthy to a certain extent, but
may be gained by almost any one who pleases. Mankind,
from education, taste, prejudice, are divided into parties,
sects, coteries, the members of which are valued, not for
their common humanity, but for the elements of difference by which that party or sect may happen to be distinguished. This is their common point of sympathy, and
the man who embodies most fully, and expresses most
strongly, the peculiarities of the party, will have reputation, will be the great man of the party. But the very
cause of his reputation cuts him off from sympathy with
the race, and he must pass into oblivion. Such are the
party men of the day, who flourish because they are party
men, and for that reason, so far as they are party men,
must fade. Such are the zealots and sectarians, whether
in politics or religion, who are distinguished by anything
which is not connected with the great interests of truth
and of duty. The possession, in an uncommon degree, of
any quality, as wit, humor, memory, will confer reputation.
It may be gained by contrivance and trick, by collusion
and bargaining.
But with glory it is not so. It has been said already
that the elements of humanity are common to all, and that
it always recognizes and responds to any vivid portrayal of
itself. We are all conscious of indefinite workings of our
11#
125
LECTURES ON MIORAL SCIENCE.
minds, of undefined and shapeless feelings, and when these
are brought out into perfect expression by the touch of
genius we are delighted. We admire, and are grateful to
the man who can give us new aspects either of nature or
of ourselves. It is the glory of all great poets and philosophers, of those who represent, and of those who analyze
nature and man, that in whatever age or country their
works may be found by man sufficiently cultivated to
understand them, they meet with a recognition and a response. This master minds alone can accomplish. Chance
has nothing to do with it. Artifice and pretence are futile
here.
In the same way military glory arises from the relation
of those who gain it to the permanent interests and universal feelings of man. What gives interest to a battle is
not that it is a theatre where brute force contends, but one
where different interests and principles are arrayed against
each other. It has often seemed to depend upon the fate
of a single battle whether liberty or despotism, civilization
or barbarism, should be prevalent in the world. When
the rights and destinies of men are thus at stake, he who
is most perfectly under the control of the master idea that
animates all, and most fuly represents it, naturally becomes the leader. It is not in him as an individual that
we are interested, it is in the principles of which he is the
representative, and of which his acts are the manifestations. If by exertion and sacrifice he cause those principles to prevail, we feel that he is the benefactor of mankind, that he is our benefactor, and the cry of admiration
and gratitude which mankind utter towards such a man is
glory.
It is, therefore, only by producing some great result that
glory can be obtained. To receive glory from mankind
126
THE DESIRE OF GLORY.
we must put ourselves in relation to them, must affect
their destinies, must make some striking exhibition of talent, or of those emotions and passions that are had in adniration among men. The man who can do this may be a
fortunate man, but he must be a great man.
Of this account of glory, in which we find the rationale
of modern hero-worship, I remark,- 1st. That as a motive
of action it can apply to but few. Few, comparatively,
can, by any possible exaggeration of self-esteem, suppose
they can produce results that shall put them in relation
with the mass of mankind. 2d. That if this glory could
be a motive to many, it would be attainable by only a
few, and so must lead to disappointment. Mankind are
so much engrossed in their own concerns that there can
exist but a certain moderate amount of admiration at the
same time. The young aspirant for fame, when he has
written or done something which he thinks extraordinary,
is surprised and vexed on looking around and finding
every man minding his own business. 3d. The opportunity for acquiring this glory often depends on causes that
are beyond the control of man. At this day Washington
could not reproduce himself. 4th. This glory depends on
success, which is not proportioned to desert. Mankind
judge by success. In the race for fame misfortune is a
crime which they never forgive. 5th. The admiration of
mankind is often given to qualities that do not deserve
it, and withheld from those that do. It is not, therefore,
always a safe guide to our conduct, or a certain criterion
of goodness, without which there can be no true glory.
Can, then, this be the glory spoken of in the Scriptures?
Or is it all an illusion? Neither. Our constitution does
not deceive us. Its tendencies need guidance, but not
eradication. This part of it is a striking indication of the
127
LECTURES QN MORAL SCIENCE.
greatness of our nature, and of its capacity of being put
into relation with vast numbers, and with great interests.
The approbation of God, and of those who judge in accordance with him, is no unsuitable motive for any. It is
such an one as an apostle thought worthy of being presented. After enumerating a long list of the worthies of
former times, he represents them as resting from their own
conflict, and watching the progress of those who have succeeded them. "Seeing, therefore," says he, "we are compassed about by so great a cloud of witnesses, let us run
with patience the race set before us." What we need,
then, is to illuminate the desire of glory by the revelations
of Christianity. Regarding ourselves, not merely as citizens of this world, but of the universe, and knowing that
God is over all, and that there is somewhere a vast assembly of the good to whom our conduct either now is or
shall be known, we may give to this principle of action
free scope.
Such is the theatre on which we are to contend for the
true glory and honor, and we are to do it in the only way
in which success is possible, "by a patient continuance in
well-doing." In this race the success of one does not prevent that of another. All may enter the lists, and all may
gain the prize.
128
LEOTURE VI.
THE AFFECTIONS. - NATURAL AND MORAL. -BENEVOLENT. - DEFENSIVE
AND PUNITIVE. - ORIGIN OF MALEVOLENT AFFECTIONS. -FORGIVENESS.
- HOW SUBJECT TO WILL.- THE INTELLECT. - LOVE OF TRUTH.
IN the last lecture we finished the consideration of
what are usually termed the desires. These have no
moral character. But desire is not excluded from the
sphere of morals. It will go with us not only as an element of the affections, but in its own proper form; for
there are really both natural and moral desires, as well as
natural and moral affections.
The desires we have considered imply no previous exercise of the moral nature, and have for their object things
without us; the moral desires imply a previous exercise
of the moral nature, and have for their object our own
moral states. A paramount desire for virtue is a virtuous
desire, and a similar desire for holiness is a holy desire.
The object of the one class of desires is that we may have
something, of the other, that we may be something. In
either case, however, the desires respect not merely the
well-being of the individual, but his capacity to minister to
others through the affections; and it is to the consideration of these that we now pass.
As the appetites have for their end a perfect body, and
the desires a perfect mind,- perfect up to that point, and
as a condition for something higher,-so the affections,
though ultimate to the individual, have, as a further end,
129
LECTURES ON MORAL SCIENCE.
a perfect society. They are that part of the constitution
of man by which he is so put in relation with his fellows
that society becomes possible.
And here we find the first difference between the affections and the desires. The object of the desires is things;
the object of the affections is sentient beings, chiefly those
that are rational and moral.
The affections differ from the desires, also, because they
are disinterested. The desires receive and appropriate
their objects to themselves. Their whole business is
appropriation, whereas the affections flow from us. We
bestow them and they appropriate nothing. There can
be no interested affection.
A third difference is, that the affections are more complex. Affection is desire, and something more. It is
impossible to have an affection for any one without having
involved in it, and a part of it, a desire for his well-being.
The affection itself, as distinguished from this desire, cannot be defined, and can be conceived only by being felt.
It is among the ultimate and highest forms in which our
humanity expresses itself.
But in analyzing the affections we are not to destroy
them. This Brown has done. He makes no such class
as the affections. The specific feeling of love, for instance,
he classes with immediate emotions, and our wish for the
happiness of those we love, with the desires. But this is
like treating of oxygen and hydrogen separately, and then
denying that there is such a thing as water. Water, which
is one thing, is neither oxygen nor hydrogen, but the two
united; and pity, which is also one thing, is neither a vivid
emotion in view of distress, nor a desire to relieve it, but
the two united; and neither can be regarded practically
in any other way.
130
THE AFFECTIONS.
But we must notice here a peculiarity of that desire
which is an element in love. As it is our own desire, its
gratification must be a source of happiness to us. As it
is a desire for the happiness of others, it must lead us to
promote that, and it is impossible that we should thus promote the happiness of others without promoting our own.
Hence, some question the possibility of disinterested benevolence. We desire, they say, the happiness of others for the
sake of our own. It is true that we are made happy in
making them so, and an admirable provision for mutual and
extended happiness it is; it is also true that we may exercise and cultivate this desire, or rather the affection of
which it is a part, as we may any other, with the knowledge that it will thus make us happy; still the desire is for
the happiness of others, and the moment it ceases to be
that, - that disinterestedly, - the affection itself is gone,
and with it the very source of our happiness. A desire
for our own happiness cannot be an element of affection,
and when, for the sake of that, we pursue towards others
such a course as affection would prompt, the whole source
and character of our happiness, if we gain any, is gone.
It may be from self-love and selfishness, but the pure happiness of affection it cannot be. The gold is become dim,
or rather dross, and the most fine gold is changed.
The affections, regarded as a whole, further differ from
the desires in being, as has been said, ultimate for man
himrnself. They refer to society; but there is nothing
within the man that is higher than they to which they
minister. So far they are ends and not means. We rest
in them. They react, indeed, on the inferior parts of the
constitution, but do not serve them in the same sense in
which they are served. Love, as involving not merely
constitutional affection, but rational choice, is the highest
131
LECTURES ON MORAL SCIENCE.
form in which our nature can manifest itself. There is in
it a synthesis of affection and of will.
From these differences it is plain that in passing to the
affections, taken as a whole, we enter another region and
group, where we find elements that are wholly new. We
come to that in the intelligent world which answers to
heat and electricity and magnetism in the physical world,
or rather to the one agent of which these may be but
the varied manifestations. Heretofore, all has been appropriation, and has looked towards self. Here self is not
forgotten in the arrangements of God, but must be by us.
The desire that enters into love retains its power of good
to us as a desire, but by thus entering loses its capability
of being abused into selfishness. As an appropriating
desire it is wholly lost. In becoming a desire for the good
of others, it becomes disinterested. Of this, the possibility, as I have said, has been doubted by some. They do
not believe that a son, knowing that he should inherit a
large estate on the death of his father, dependent on his
assiduity, could attend upon and cheer him through his
final sickness purely from affection. They are in the same
position as the heathen, who cdnot conceive that the
missionaries should come with the simple object of doing
them good, whereas the whole glory of the missionary
work is in its unselfishness. When that departs, it is shorn
of the locks of its strength, and becomes like any other
cause. But in this structure and action of affection we
simply find the paradox of our Saviour that he who would
find his life must lose it. That is not peculiar to his religion. It has its basis in our nature. It is the condition on
which any higher life of the affections is to be found. It
is by losing all thought of himself that a man finds his
own higher self. The ultimate happiness and good for
132
THE AFFECTIONS.
man is something more than the happiness from desire, as
found in affection. That is there the inferior and weaker
element. It is from a union in sympathy, of which desire
knows nothing, from a mutual love; it is in a glow, and
ardor, and exultation ineffable in view of the high powers
and qualities of other beings to whom we are united by
an unalterable affection, - an affection springing from the
very depths of our rational and voluntary nature, and
through which we find relationship and kindred dearer
than any other. Here again the Saviour understood our
nature, and hence condensed all the natural relationships
into one to express that of moral affinity. "For whosoever," said he, "shall do the will of God, the same is my
brother, and sister, and mother." The blessedness from a
sympathy and love where there is perfect moral complacenlcy, who can estimate? Who can estimate the repercussion and multiplication of joy when each one shall not
only have joy in himself, but shall also rejoice with all
that do rejoice? Hiow shall the whole principle and
method of selfishness be reversed, when, instead of looking
on his own things, every man shall look also on the things
of others, not with envy or jealousy, but with the greater
delight as the gifts and endowments are greater, and shall
feel that he owns them all in a far higher sense than he
who can enjoy it owns the landscape!
In affection it is the union in sympathy that is the electric element, and this may pervade society as if it were a
living organism, and so that whatever is felt by one shall
be felt by all. From what we see of the power of sympathy in large bodies of men, in nations engaged in a common cause, where there is yet much selfishness, and the
means of communication are imperfect, we may imagine
what it would be if there were no selfishness and the
12
133
LECTURES ON MIORAL SCIENCE.
means of communication were complete. With these conditions, the "joy in the presence of the angels of God
over one sinner that repenteth" would follow, of course.
Not only do the affections, as has been said, point
towards society, but they are the only social element. The
appetites and the desires both appropriate. They are not,
in their own nature, and of necessity, selfish; but they
have a primary reference to self, and become selfish when
they so act as to encroach upon the sphere of the faculties
above them. If a man so indulge his appetites as to
encroach upon his desires, -if the love of eating overmaster and dwarf the desire for knowledge, -there is selfishness as well as sensuality in the act, because the man
dwarfs his higher nature, and so unfits himself for the good
he might do to others. So, too, though the desires, acting
within their own sphere, are merely manifestations of our
nature having reference to self, but not selfish; yet if they
encroach upon the sphere of the affections, they immediately become selfish, and it is one of the common and
prominent forms of selfishness for them to do this. With
only appetite and desire, the whole object of man would
be appropriation to himself, and he would use his fellowmen as things, simply for his own convenience. Men would
care no more for each other than the player does for his
nine-pins. Association there might be, but no society; and
the association would have about it no charm, no beauty,
no warmth, nothing disinterested or noble. But let now
the affections come in; let friendship, and gratitude, and
pity; let sympathy and love in its various forms, as conjugal, filial, and fraternal affection, appear, and they make a
new world. They are like the angels from heaven descending among men. They come, and mere forms, and conventionalisms, and hypocrisies and overreachings, give
134
THE AFFECTIONS DISINTERESTED.
place, and disappear like birds of night before the light
of day.
That the affections are the only social element, it is
desirable to notice, because it shows us precisely what we
are to cultivate to make society perfect; and also how it is
chiefly liable to be corrupted, or rather perverted. This is
by the coming in of the desires where the affections ought
to rule. The affections, as I have said, cannot be interested.
A true friendship cannot be so, and hence its beauty, - a
beauty scarcely paralleled on earth. But if we suppose
those acts which seemed to be prompted by friendship, to
be really prompted, not by affection going out towards the
person, but by the desire of some benefit from him, the
beauty will vanish in a moment, and contempt and detestation will take the place of complacency and admiration In married life, and in all preliminaries to it, there
is beauty as the affection is pure, not only from sensuality,
but from all desire of property, or of any incidental advantage. There is an expression employed by somne, - that
of using one's fiiends,- that was always offensive to me.
The displacement in society of affection by desire is bad
enough, but the shameless avowal of it is worse. Here is
a chief ground of the hypocrisy noticed in connection with
the desire of esteem. Nothing can be more annoying or
chilling than to be in a community where there is a universal tendency to gratify some form of desire under the
profession and appearance of affection, and especially to
boast of success in this as an evidence of smartness and
of a knowledge of human nature. This it is that gives to
fashionable lifb, when the people who are in it understand
each other, as they generally do, its heartlessness, and lays
it open to the shafts of satire. Let its polished but meag,re
conventionalisms be filled out with a hearty affection, and
135
LECTURES ON MORAL SCIENCE.
it would be like the resurrection and free motion of a
corpse that had simply been galvanized. The same is true
in all the relations of life. The corruption is, that appetite
and desire, and so sensuality and selfishness, have usurped
the place of the affections; and the great thing needed in
society is that these should assume their due prominence,
and rule in their own sphere.
In what has been said hitherto no distinction has been
made between the natural and the moral affections. That
distinction we must now draw, for in strictness it is only
the natural affections that should be spoken of here.
The character of an affection is determined by its origin
and its object. The natural affections are those that spring
up impulsively as do the appetites and the desires, and are
such as we share in kind with the animals. They do not
spring from the moral nature, and have no regard to the
moral character of their object. They have, therefore, no
moral character in themselves, but, like the appetites and
the desires, are purely instrumental, and are good and
evil solely as they are controlled. They are good in their
place, and for the purpose for which they were intended,
but are not morally good, and do not become so by being
brought under moral control. The moral affections spring
from the moral nature; and it is upon moral beings, as
such, that they rest.
This distinction seems plain, but may, perhaps, be made
more so by a reference to the language of the Scriptures.
In them the term "Heart" is used to signify the affections, but not the natural affections. In the expression,
"My son, give me thy heart," we feel at once that, while
the affections are meant, there is yet an entire exclusion of
anything like the natural affections. That expression carries us at once into a region that is wholly moral and free,
136
THE HIGHEST FORM OF ACTIVITY.
and when God is the object of affection there is in it the
highest possible form of activity. A supreme love, as that
of God must be, if it be at all, involves the choice of a supreme end, and that was shown in the second lecture to be
the highest act of a rational being- the outgrowth of his
whole personal activity. From what was said at one point
it might be supposed that the will, as distinguished from
the affections, would be the highest, but in this love there
is a coalescence of will and affection such that the love
may be said to be the two united. There is in it a rational
preference which belongs to the will as free; there is in
it benevolence and the highest complacency and delight.
These are not there as separate elements, more than the
ultimate elements of the flower, the oxygen and hydrogen and carbon, are separate in that. They tend to make
up the one love, which, as the joint product of the highest faculties of man, thus becomes the one "consummate
flower" of his existence. Not unlike is it to the flower of
those plants which put forth but a single one at the top,
and which is the product and highest expression of their
whole life.
But while the line between the natural and the moral
affections is thus theoretically distinct, it is many times
both difficult and important to distinguish them practically.
It is difficult, because they so conspire together, and
seem to permeate each other. They are often, we may
say generally, in exercise at the same time, and with the
same person for their object, and the whole result becomes
so blended into one as to be inseparable. When the two
conspire there is a perfect complacency and satisfaction,
but we cannot tell how much to attribute to each; when
12*
137
LECTURES ON MORAL SCIENCE.
they conflict it is often difficult to say how far each should
prevail.
It may also be important to separate the natural from
the moral affections. It may not. What nature gives us
together we may receive together. We may eat the pudding with no attempt, even in thought, to separate the
sugar that pervades every particle of it from that which
forms its basis; and that is the pleasantest way. Still, if
we would judge accurately or even fairly of men, the line
which separates the two forms of the affections must be
drawn. The reason is that the natural affections are liable
to be mistaken for moral character. In all that pertains
to the natural affections the differences of endowment by
nature, and with no reference to moral character, are as
great as they are with reference to the intellect, or strength
and beauty of body. In some these natural endowments
are rich and free and beautiful in their spontaneous action.
Such are said to be, and they are, amiable. Others are the
reverse of this. It is no fault of theirs; it is an infelicity. One is the rose, and the other the nettle; one is the
smooth, and the other the rough-barked tree; and nature
has made the difference. Still, it is not uncommon to find
the richest gifts of the natural affections as well as of the
intellect associated with the deepest moral corruption.
To this there seem to be even some special tendencies. It
is the smooth-barked hickory that bears bitter nuts. Nor,
on the other hand, is it uncommon to find from those less
happy in natural endowments moral manifestations that
surprise and delight us the more on that very account.
They are as the good nuts from the shag-bark hickory;
they are as the beautiful flowers from the prickly and angular cactus.
While, therefore, we recognize all there is in this part
138
THE AFFECTIONS CLASSIFIED.
of our nature of beauty, and of desirableness for social
life, we are not to confound gifts with virtues, nor natural
kindness of heart and amenity of manners with moral
principle. We admire the natural affections, we regard
any great lack of them as a deformity, we are apt to censure it as if it must have had a moral origin; still, if the
moral nature be withdrawn, there remains in them but the
foreshadowing and prophecy of something yet higher.
The distinction now made is indispensable; but, in treating of what are called the natural affections in man, we are
not to suppose we are treating of them alone. The light
of the moral affections constantly shines through them,
and gives them a radiance not their own. In what shall
be said further of the affections, this distinction, therefore,
need not be particularly regarded.
By most writers the affections are divided into the benevolent and the malevolent. But the term malevolent
is unfortunate. Perhaps it would be better to designate
them with reference to their end rather than their origin, and divide them into those intended to make others
happy, and those for self-defence and punishment. We
should then have for one class what we must still call the
benevolent affections, and for the other the defensive and
punitive. Affections strictly malevolent are not to be
presumed, but self-defence and punishment are each necessary and proper, and we might expect there would be
affections that should indicate these and support us in
them.
In inquiring after the number of the original benevolent
affections, for these come first in order, we are to be guided
in the same way as when we were inquiring respecting the
desires. Consciousness must be the ultimate test; but if
we can ascertain what affections would be necessary to the
139
LECTURES ON MORAL SCIENCE.
upbuilding of a perfect society, we may be sure we shall
find just those, and no more. The provisions of God are
always adequate for their end, with nothing superfluous.
The first will, of course, be those that belong to the
family as a divine institution into which man is born. He
does not originate it, but is born into it. Here we have
the conjugal, parental, filial, and maternal affections, and
where these exist in their purity and proper power, a family is as the garden of God. It is worthy of being his
institution. It is the centre of the affections, the home,
the sphere of the purest and best earthly happiness, and
the germ and source of all civil institutions. But besides
these we have "the special and distinguishing affection of
man towards woman, and of woman towards man, which
tends to the conjugal union. This is expressed by the
word'love, without any epithet." We have also sympathy, pity, gratitude, friendship, patriotism, and general benevolence, or philanthropy. These may all be included
under the word love, as their opposites may be under the
word resentment.
Of the benevolent affections generally it may be said
that the pleasure already noticed as connected with them
may be regarded as an expression of the approbation of
God, and as an invitation to us to cherish them. It expresses the wish of Him who made us that these affections
should prevail, and evinces his benevolence, since they
are at once happiness in those who exercise them, and
productive of happiness in those towards whom they are
exercised.
We are also fond of seeing excess in these affections
rather than deficiency; and if duty be called upon to
control them, we choose it should be for restraint and
repression, rather than excitement. It is the excess of
140
SHADES OF AFFECTION.
these affections, the preponderance of the natural element
over that which is moral, which gives rise to what are
called amiable weaknesses. Through these our respect for
a man is diminished more than our love. The tenderness
of a father for his child may be a little laughable, yet we
easily forgive it, and prefer it to the least want of affection.
i We have thus a double provision for the encouragement
of these affections, - their effect upon our own happiness,
and the sympathy of others.
The terms above given may indicate sufficiently the
various forms of affection to enable us to speak of them
intelligibly, but those affections are constantly differing
from each other and from themselves as their objects
differ. What it is to love can be known only by loving,
and to appreciate the different shades of affection we must
ourselves have felt its nice and varying adjustment to its
varying objects. A feeling of responsibility, of anxiety,
which is a mixture of hope and fear, of protection and
of peculiar tenderness, is blended with parental affection.
Filial affection is modified by gratitude, confidence, respect, and reverence. If the graver and sterner virtues
enter largely into the character of our friend, we feel for
him more of respect; if he be of the softer mood our
affection partakes of that character.
Of the particular benevolent affections mentioned it
would be pleasant to treat particularly and at large, but
they are so far of one general character and object that
that will not be requisite. Like the appetites and the
desires, they are to be controlled with reference to their
end, and will be most for the happiness of the individual
when they are so controlled as to build up the most perfect home, and the most perfect civil society.
I will only add, that as society originates, and finds its
141
LECTURES ON MORAL SCIENCE.
beauty and blessedness in these affections, so it will react
upon them. That form of society and those habits of life
are, therefore, to be preferred which give to these the best
theatre and widest scope. Especially is it to be said that
the breaking up of the home for any system of communism or socialism must be equally opposed to the intentions of God and to the highest happiness of man.
But besides the benevolent affections there are, as has
been said, those that have been called malevolent. Concerning the origin and character of these there has been
great diversity of view. There still is. But the part they
have played in the history of the race is so conspicuous,
and they are so difficult of control, that they ought to be
well understood.
Concerning these, two remarks may be made, the opposite of those made concerning the benevolent affections.
The first is, that this class of affections, at least so far as
they are malevolent, are painful to those who exercise
them, thus indicating the will of God that they should not
be indulged in.
The second remark is, that mankind are pleased to see
these passions repressed and moderated below their natural standard rather than suffered to rise above it.
As, then, we found a double provision for the encouragement of the benevolent affections,- our own satisfaction
and the sympathy of others, - so now we find a double
provision to repress these opposite affections, - our own
pain and the disapprobation of others.
Since the time of Butler a distinction has been made
between "sudden," as he called it, or, as it has since been
called, instinctive resentment, and that which is deliberate.
The first is the guard appointed by nature against any
sudden attack. It is the assertion by whatever is, of its
142
FOUNDATION OF RESENTMENT.
right to be, and confers a promptness and energy whichreason could never bestow. This is purely the work of
nature, and cannot be wrong. It is only concerning deliberate resentment that there can be a question.
And here my wish is, by tracing its origin, to vindicate
this part of our constitution from the charge of anything
malevolent, properly so called, and to show its propriety.
This will give us the key to its proper use.
Let us then suppose two moral beings, one perfectly
good, the other perfectly bad, to meet together. It is
clear that they could have no coincident wishes, but would
naturally array themselves against each other. "What
communion hath light with darkness?" If now the evil
being should exert a particular act of injury towards the
good, what would be the feeling of the latter? It could
not be the same that he would have towards a being perfectly good. What will you call that necessary opposition,
that sense of repugnance, of dislike, of condemnation, of
abhorrence even, which the good being could not but feel?
It is this opposition of virtue to vice, of holiness to sin,
that is the proper foundation of resentment, and that becomes the only resentment that is justifiable when vice
exerts itself towards us in a definite act. In this view of
it, resentment is nothing more than a sense of ill-desert
where it really exists, and a desire to punish it so far as is
necessary; and so far from being opposed to goodness, in
the wide and proper sense of that word, it is a necessary
part of it. The hatred of vice is the opposite pole to the
love of virtue, and the positive cannot be evolved without
the negative side. Of necessity, the strength of the one
is the measure of that of the other. Moral purity, virtue,
holiness, whatever we may choose to call it, is not a mere
passive, undiscriminating quality. Nothing can be more
143
LECTURES ON MORAL SCIENCE.
positive, active, and uncompromising. Against whatever
is opposed to it, it arrays itself in a conflict that can know
of no cessation and of no compromise till one or the other
is completely triumphant; and this opposition cannot be
malevolent, since precisely as it prevails happiness is extended, precisely as it fails misery bears sway. If this
were not so, there could be nothing venerable or awful
about goodness. It would not command our respect, or
be worthy of the throne of the universe. This is what is
termed in the Scriptures the anger of God, without malice,
without revenge, without respect of persons except as
good or evil; and it will be the misery of those who shall
be finally opposed to God, that they will be opposed to
Infinite Goodness, and that Infinite Goodness will be
opposed to them, not because it is malevolent, but because it is Infinite Goodness. Here we find the source of
all penal law. Without this, there could be no security,
punishment, or redress. It is this feeling, which, on the
perception of wickedness and ill-desert, if the injury be
ours, we term resentnent; if it be upon others, indignation; but the principle is the same, and is entirely different
from malevolence. This was no part of the human constitution as made by God.
When we speak of the opposition of virtue and vice, it
will be remembered that these are mere abstractions.
Strictly speaking, there are only virtuous and vicious persons, and hence the punishment of vice must involve the
infliction of personal misery, though without malevolence.
But it will be asked, If there be in man no other malevolent principle than this, how shall we account for the
jealousy, the envy, the hatred, the malice and revenge that
fill and disfigure the earth? They may, I think, all be
traced to the perverting influence of selfishness on the
144
JEALOUSY AND ENVY.
original and natural principle of resentment. To be satisfied of this, a brief reference to each will suffice.
Jealousy is an affection which has direct and sole reference to self. We are jealous of no one who is not or may
not be a rival. It is when interests are likely to conflict
that selfishness, without provocation, stirs up that form of
ill-will which we call jealousy. The same may be said of
envy. Indeed, envy and jealousy are the same affection
towards persons differently situated. That which is jealousy towards those who we fear may surpass us, becomes
envy when once we are fairly distanced in the competition. In witnessing a contest, we feel no envy. Envy is
the dislike of those who are above us because they are
above us, and a desire to pull them down to our own level.
It is, therefore, directly to selfishness that these two evil
affections may be traced.
But jealousy and envy are apt to become settled hatred.
Hear those who are, or have been competitors, speak of
each other, and you will find the reason. You will find
that they impute the success of their opponent to unfairness in him or others, -to some cause which will justify
them in showing resentment. Their self-estimation will
not permit them to think otherwise. With her jaundiced
eye selfishness can convert even the excellences of others
into faults, and then, having something that she supposes
she can fairly blame, she usurps the place of conscience,
and calls upon resentment, which in this unholy alliance
becomes malice, to pursue them. Hatred does not spring
up naturally from the relations in which we find ourselves
in society, as do the natural affections, but requires as its
condition some injury real or supposed. So with revenge.
By its very nature and definition it implies previous injury,
and it is nothing more than the natural feeling of resent 13
145
- LECTURES ON MORAL SCIENCE.
ment exaggerated by selfishness, and abused into a settled
and inveterate passion.
The extent of these passions, and the slight occasions
on which they are permitted, are indeed wonderful, but
they may all be traced to the combination of selfishness
with the natural and necessary principle of resentment.
If we distinguish between actions that are simply injurious
but not malevolent, as when a robber plunders another,
not from any hatred of him, but from a love of his money;
and if we make due allowance for the operation of a perverted and perverting selfishness on the natural feeling of
resentment, we may see how far man may be said to have
originally affections that should be called malevolent.
Here, as elsewhere, evil is from the perversion of that
which was good. That part of our constitution from the
perversion of which these affections arise, we vindicate.
It is essential to goodness itself. It guards our highest
interests; it is the basis of penal law, keeping crime and
tyranny lurking in their lair; and no character which cannot, and, if need be, will not reveal itself as opposed with
the force of the whole being to moral evil, can command
our respect.
But while we thus vindicate this part of our constitution
as it was originally given, we utterly condemn all jealousy
and envy, all hatred, malice and revenge. They are not a
part of our original constitution, and were never made by
God. Jealousy and envy are not only among the basest,
but are the meanest, of the passions. They are indulged
in only by those who are conscious of inferiority, and are
not only malignant, but are a confession of that degrading
consciousness. These, as well as malice and envy, bring
their own punishment with them. They bring it in the
disquiet which they necessarily cause to their possessor
146
FORGIVENESS.
and in the detestation with which they are viewed by
mankind. We cannot, therefore, too sedulously stifle these
double curses, curses upon ourselves and others, and
which have so filled the earth with discord and misery.
In opposition to these, I cannot forbear to mention that
beautiful trait of Christian morals, the forgiveness of injuries. How infinitely superior is this to all recrimination!
Revenge places us, at best, upon a level with those who
have injured us,- forgiveness elevates us far above them.
And then how fitted is it to the condition of man! If,
as all experience shows, there must be mutual forbearance
in the end, why not exercise it before suffering the miseries of mutual recrimination? If we all need forgiveness
firom God, how suitable that we should forgive each other!
To make an offence unpardonable simply because it is
against ourselves, is the arrogance and-blindness of selfishness, and involves a principle that would preclude all forgiveness. Forgiveness and placability are not meanness
or pusillanimity,- they are that attitude of humanity in
which it most resembles God.
In considering the affections in connect'n with morals,
we next inquire how far they are subject to the will.
There are those who suppose that the affections and
passions are enkindled and drawn from us by a fixed law,
as electricity flashes from one cloud to another, and that
we are therefore not responsible for them. But the voice
of mankind is that men are responsible for ftheir feelings
through the whole range of the emotive nature, as well as
for their actions. They judge that men can govern their
passions, not only by restraining those external acts to
which passion would excite them, but also by moderating
and subduing the feeling itself. This is correct. Men are
responsible not only for the feeling they have, but also for
147
LECTURES ON MORAL SCIENCE.
not having the feelings they lack; and yet no man can, by
any direct act of the will, cause any one feeling, affection,
or passion, to exist. Throughout its whole range the ermotive part of our nature is excited by an object adapted to
excite it, and not by a direct act of the will. No man can
feel pity simply by willing to feel it. I-Ie must have a view
of poverty, sickness, distress, in some form, and it will arise
in view of these. No man can feel gratitude except in
view of a benefactor. No man can feel love or respect
for one of whose character he is ignorant. These must
arise in view of excellence real or supposed. Not on a
direct act of the will, nor on the object as it is in itself,
but on the object as viewed by the mind, will the feeling
depend. A lover may suppose that he sees perfection
where another would not see it, and where possibly it does
not exist, but the feeling will be the same. Here the feelings and affections called moral are governed by the same
laws as those that are not. Were a person commanded to
feel the emotion of beauty, as he may be, and is, to feel
the affection of love, he might shut his eyes and say that
he had no control over his emotions, but not thus could he
escape the obligation. Before he could do that, he would
be required to seek out some fair face, or beautiful form,
or exquisite work of art, or to find his way to some commanding eminence whence he might cast his eye over
mountain and valley, the cultivated field and the winding
stream, and then if he should not feel the emotion of
beauty he might be absolved from the obligation.
But as men have different degrees of feeling on viewing
the same object, it may be asked, What should be thought
of one who should give his best attention to an object
adapted to produce a given feeling, and not have it?
What should we think of a man who could thus see dis
148
HARDNESS OF HEART.
tress but feel no compassion? That would depend on his
previous history. If he had gone through no hardening
process, I do not see that he would be worthy of censure.
Such a case we should regard as unnatural, as monstrous,
as we should a natural deformity, but it would simply indicate the want of an original capacity. To most persons
the feeling of pity on meeting suddenly with a scene of
distress is as unavoidable as the feeling of surprise on
meeting with one that was unexpected; but as the spontaneous presence of a feeling without the intervention of
the will is not a virtue, so its absence, where the susceptibility is wanting, is not a fault. This, however, is not the
hardness of heart which mankind condemn. That comes
from the over-mastering power of some cherished and
selfish passion. A man who has given himself up to the
love of gold, when he pays his visits to his poor tenants on
thevery day their rents become due, can see nothing, and
hear nothing, and think of nothing but money. His mind
is so absorbed by that, that there is no room for any other
feeling. It is not that he has no susceptibility, but that an
absorbing selfishness has closed up the avenues to his
heart. Seeing, he sees not. His finer susceptibilities fall
into desuetude; a current gets its set in his soul which
undermines and washes away everything beautiful, and
then, indeed, by wrong action long indulged, comes that
hardness of heart which the world justly condemns. It is
by a process like this that the priests and the Levites are
formed who pass by on the other side. They know there
is misery in the world, but they not only do not seek it
out, they avoid it, lest it should disturb their selfish quiet.
In what I have said hitherto, the object which was to
awaken feeling has been supposed present to the senses,
but this is not necessary. Mental representation and
13*
149
LECTURES ON MORAL SCIENCE.
thought will do as well. It is thought that is the condition of feeling, and governs it. As we think of a man
so do we feel towards him. But that the mind has a control over the thoughts, can dismiss some, and dwell upon
such as it prefers, can study one subject and not another,
none will deny. If, then, any choose to let vain thoughts
lodge within them, they must not complain that their affections are disordered; and if any choose to regulate their
thoughts, they will thus indirectly regulate their affections
also. Our control over the affections is, therefore, though
indirect, yet efficient, and such as to render us fully accountable for them. A man can determine the kind of
affections he shall have in the same way that he can the
kind of knowledge he shall have, or the kind of feelings
that shall be produced by the food which he eats. After
he has taken food the will has no power to modify the
effects, but by his power over the taking of that the man
has indirectly a power over his bodily feelings and general
health. A man may disorder his affections as he may his
body, and be as responsible for the one as for the other,
and that not by directly willing it, but by doing that
which he either knows, or might know, would produce
that effect.
But feeling, it may be said, reacts upon thought, and in
turn governs it. True, but this is not the natural order.
If we have become subject to a wrong governing passion
everything will be seen in the light of that, and no person
who shall stand in the way of the gratification of that
passion can be viewed as he really is. Here we find the
difference in their effects between a governing principle
that is legitimate and one that is not. A legitimate principle can create no interest in the mind to prevent it firom
viewing everything as it is. It originates in light, and
150
AFFECTIONS CAN BE CONTROLLED.
seeks the light. But no man can adopt a supreme end that
is wrong, and be willing to come to the light. Iaving
adopted his end, whatever shall be opposed to the attainment of that will be viewed with prejudice and regarded
with hostility. This would be required by the demand of
every rational nature for consistency with itself, and in this
we find the origin of all criminal prejudices, and of much
of the malignity there is in the world.
The affections, then, and the feelings generally, being
thus controlled through the power of attention, how intimate must be the connection between the character and
happiness of a man and the objects to which he habitually
turns his thoughts! It is not only habits and associations
of thought that are thus formed, but also of feeling; and
thus the whole character is reached. He whose course of
reading and meditation presents only what is pure, and
generous, and noble, lives in a world wholly different from
that of him who pursues an opposite course.
It is to be observed, too, that that system, aside from
any consideration of its truth, which presents the noblest
objects to the intellect, must also be most favorable to
high and noble feeling. Hence atheism, which takes away
the conception of an infinite and perfect God, or rather
renders it inoperative, must wither the feelings and debase
the character, and any system which shall present the
Godhead as shorn of any possible perfection must have
proportionately the same effect. Hence the vices and
degradations of heathenism are naturally associated with
idolatries. It was when men "changed the glory of the
uncorruptible God into an image made like corruptible
man" that "God gave them up unto vile affections."
This doctrine that we are responsible for the affections,
and particularly for the natural affections, has special need
151
LECTURES ON MORAL SCIENCE.
of enforcement at the present time. Always there have
been those'who have justified lawlessness here on the pretence that the feelings were irresistible. They have even
gloried in it as indicating the warmth and richness of a
nature of superior endowments. It was not for them to
be cramped by the rigidity of rules. How could feeling
be brought under law? What had they to do with the
sternness of principle? That they left to colder natures.
They have had no conception of desire sanctified by affection, and strengthened and regulated by principle; but
under the pretence of affection, in the sacred name of the
heart, they have cherished selfish and licentious desires;
and, as wolves in sheep's clothing, have crept into and
desolated the fold of domestic peace. But never, perhaps,
has this been so common as now. Now, on the ground of
affinity, or the want of it, the most solemn vows are violated. Husbands desert their wives, and wives their husbands, and even their children, and doctrines are openly
taught, claiming sanction from the spiritual world, that
would subvert the most sacred institutions of society, corrupting it at its fountain-head, and that would obscure
and defile those pure relationships in which its beauty and
strength now abide. As constituted by God, with its origin in the family, society is a soil congenial to the natural
affections, and, unless checked by the selfish passions, they
will spring up. These selfish passions men can repress.
They can have some degree of consideration in forming
the conjugal union; they can substitute principle for passion as a controlling power, and can dwell on those aspects
of character that would excite affection. They can coiltrol many circumstances and conditions that bear on the
affections, and will be sure to foster them; they can perform the external acts which the affections require, and
152
THE INTELLECT.
form habits of them, and these will react on the feelings;
they can regard the higher interests of those with whom
they are associated, and cause the LAW OF LOVE to take
the place of those capricious emotions which have nothing
in common with it but the name. They can thus vindicate the supremacy of the moral nature, and, instead of a
sweltering and chaotic mass of moral corruption, tending
to a corruption still deeper, can cause society to present
the order and beauty of the planetary spheres.
In our division of what were called the instrumental
powers, the powers that are to be governed, that are for
an end beyond themselves, we made one class of those
that indicate ends, and another of those in the light of
which ends are pursued. The first class we have now
considered, and a few words will suffice for the second.
These may all be comprised under the one term Intellect; but will include only those faculties and operations
of intellect that may be modified by the will. It will
include all those faculties by which we arrive at truth by a
process, and will exclude those that are intuitive.
In a system of psychology it would behove us to consider these powers before those of emotion, since something
must be known before anything could be felt. But we
are now considering ends, and the initiatory step towards
an end is not, as has been said, in the intellect, but in
some tendency or craving, some feeling of want or appreliension of excellence. For this the intellect is indeed a
condition; but it seemed more accurate to begin with our
fundamental conception, and the powers which give us
that, and then to regard the intellect as simply instrumental. This, however, is a mere question of arrangemnent, and is not particularly important. What is important is that we should apprehend fully the connection of
153
LECTURES ON MORAL SCIENCE.
truth with the rational pursuit of ends, and our responsibility for a knowledge of the truth.
That the intellect, as now defined, is simply an instrument, there can be no doubt. It is difficult for us to separate wholly the operation of knowing from ulterior results;
but when this is done we see that it cannot be an end in
itself. Knowing is in order to feeling and action, and
without these would be wholly barren. The law of the
intellect, therefore, like that of the powers already considered, will be from its end. That end, as we now contemplate it, is not knowledge to be acquired promiscuously under the stimulus of curiosity, but practical truth,
as the first condition of wisdom. For this the intellect
was given. It was intended for this, as the eye for seeing,
and the true dispositions required in the conduct of it are
earnestness and candor.
Of these, earnestness will secure that self-denying labor,
that careflul analysis, and patient induction, and comprehensive research, which the Scriptures imply when they
say, "buy the truth and sell it not;" and candor will
secure us against all biases fiom interest, and, as was said
under the affections, from our having already chosen a
wrong supreme end. This is the same as that singleness
of eye spoken of by our Saviour, through which, if a man
has it, "his whole body shall be full of light." It is a disposition -and this shows the philosophy of what was said
by our Saviour, and cannot be too strongly enforced -a
disposition which is impossible to any man who has (chosen
a supreme end that is wrong. On some points hlie may be
candid, but not il reference to those persons and things
which would thwart him in the attainment of that end.
No man can be wholly candid who has not chosen the
right supreme end, and so has no interest that he conceives
154
RESPONSIBILITY FOR OPINIONS.
to be supreme, to be otherwise. It is not that candor cannot exist where it is opposed to interest, but only to that
which is regarded as supreme. In this case a man cannot
consistently come to the light. To do so would be death
to him in that which he counts his dearest interest, and so
his very life. But he who has chosen the true supreme
end and pursues it in simplicity, mtvt see all things truly.
There can be no refraction in his mental vision, and his
whole body will be full of light. This is the only position
we can take in which the light that God sends will not be
refracted. Without this we shall see some things falsely;
more or less we shall "walk in a vain show."
On this subject we concede that there are laws of evidence. Nay, this is the very thing we assert, and it is just
because there are such laws that we hold men responsible
for their opinions. Without them they could not be. If
there were no certain road by which a man could reach a
given place, he would not be responsible for not getting
there. But if there were such a road, and he should be
too careless and self-confident to inquire for it, or should
think it too difficult or disagreeable for him to travel, he
would be responsible. So here. Truth is one. It corresponds to the mind as light to the eye. It was intended to
be seen, and if the laws of evidence, the fixed condition
of our receiving it, be fully complied with, it will be seen
as in a pure white light, and, so seen, "will make men
free." If not, the mind is wrongly constituted in its relation to the objects of knowledge, and the constitution of
man is hopeless. What should we think of a man who
should hold a prism before his eyes, or shut himself in a
room with windows of colored glass, and then complain
that he could not see objects in their true color, because
there were fixed laws and conditions of vision? Let there
155
LECTURES ON MORAL SCIENCE.
but be earnestness and candor, and nothing can prevent the
truth from being both seen and received. But for earnestness and candor we are responsible. This none will deny
who admit of responsibility at all.
The truth is, we are so endowed and so placed as to be
capable, not of all knowledge, nor of freedom from mistakes, but of knowing the truth so far as it bears practically upon our highest interests. This we cannot do by
any direct act of will, but through fixed conditions, which
will ensure it, and to comply with these conditions is
among our very highest and most sacred duties. To love
the truth is here the first and great commandment; to tell
the truth, which is like it, and a corollary from it, is only
the second. It is, however, a duty that has been too much
overlooked. In our current treatises on morals, truthfulness has had a large place, while this primal and higher
duty of knowing the truth has been scarcely noticed. To
this the time permits me simply to give, as I have now
done, what seems to me its true place, and in doing so I
bring to a close the consideration of those powers which
require to be governed, and whose chief end is out of and
beyond themselves.
156
LECTURE VII.
THE MORAL NATURE. - REASON. - IDEAS OF DIFFERENT ORDERS. - HAVE
AN ORDER OF DEVELOPMENT. - FREE-WILL. - PERSONALITY.- ACTION
TO WHICH RESPONSIBILITY ATTACHES. -ALL MORAL PHENOMENA IN
CONNECTION WITH THE CHOICE OF A SUPREME END.-CONSCIENCE. THE MORAL NATURE DOUBLE. - THE HIGHEST GOOD. - COINCIDENCE OF
NATURAL AND REVEALED LAW.
INSTINCTS, appetites, desires, natural affections, intellect,
-mere intellect, or understanding, -these are all subordinate and instrumental. They are not, for man, governing powers; and however they may be, in whole or in
part, a condition for the moral nature, they are no part of
it, and may be conceived of as acting wholly without it.
In passing upward in the scale of being we reach, as I
have said, points of transition where there is no longer
merely gradation, but a leap, and the introduction of
something wholly new. We come to a difference, not in
degree, but in kind. So we find organization in a variety
of forms, and in great perfection, before a nervous system
is introduced. That, as endowed with sensation, is wholly
new. It supposes antecedent and auxiliary organization
into which it may be put, of which it may take possession,
and which may minister to its ends. There is much in
every vegetable that simulates, and seems to anticipate a
nervous system, but it is not there, and when, with its
filaments and centres, it first pervaded an organization
adapted to it, and responded consciously to the stimulus
of the external world, it was as if there had been a new
14 157
LECTURES ON MORAL SCIENCE.
star set in the heavens. There was a new order begun;
there was an animal. From that point all was gradation till a moral nature was introduced. Then there was
another leap. Then man was made, and, if a little lower
than the angels, yet in the image of God. Then organization became the abode and instrument of a spiritual and
responsible being. It is concerning this moral nature that
we are now to inquire.
And here I observe that whoever can tell what that is
that is put into the animal nature and uses it, as a nervous
system might be put into a vegetable organization and use
it, so that there shall be a person of which the animal
nature shall become the subordinate constituent, can tell
what the moral nature is, for man is no further a person
than as he is moral. Here it is that we find the ground
and necessity of that threefold division of man into body,
soul, and spirit, which the Scriptures seem to recognize,
and which philosophy will be compelled to adopt. To
this division our use of terms conforms but imperfectly;
but as thus used the soul will include those powers of intelligence which we share in common with the brutes, and
the spirit those higher powers which we now seek, and in
which personality is found.
It was said in the third lecture that the special difference between man and all that is below him is, that he
chooses his own end, or rather that he may either choose
or reject the end for which God made him. If this be so,
the powers to be added must be those in virtue of which
he does this. According to distinctions already made, they
must be directive and not instrumental. The nature thus
added must be ultimate, that is, it must minister to nothing within the constitution above itself.
Nor will the addition of such powers be a slight step
158
REASON.
upward. The transition from dead to vitalized matter
cannot be greater. It is as a new morn risen upon the
high noon of animal existence. In the powers thus given
must be those for comprehension, for control, for wisdom,
in distinction from mere devices and cunning. A brute
has not merely instinct, but some degree of understanding, by which it may vary means, and adapt devices often
showing much cunning, for the attainment of ends; but
the end itself of an animal is the outgrowth of its organization, and admits of no alternative. But man is capable
of knowing the difference between good and evil, and of
choosing between ends that he may adopt as ultimate and
supreme. He can either adopt as supreme the end proposed by reason and sanctioned by conscience, or follow
his propensities. Hle can either serve "the flesh" or " the
spirit," and one of these he must do. It is in his manifoldness, for which the capacity is thus given, that his greatness is seen. It is in the choice of the right end that there
is the supreme wisdom, even though the best means may
not be chosen; while, if the wrong end be chosen, though
there be the utmost skill in the use of means, there is yet
supreme folly. So is it that we have singleness of eye; so
that the open and straight path of wisdom differs from the
tortuous course of the serpent; so that " the children of
this world are wiser in their generation than the children
of lig,ht."
What, then, are the powers needed that man may choose
his own ends? How can there be, not only impulses from
behind that may impel, but also an end before that may
be yielded to and adopted? What are the elements of
personality, and the a priori conditions of a moral act?
Of the powers thus sought the first is Reason. In com mon language man is distinguished from the animals by
159
LECTURES ON MORAL SCIENCE.
saying that he has reason. Of this the correctness is not
to be questioned if we mean by it that which gives origin
to the word rational, rather than to the word reasoning.
No one supposes animals to be rational, while many con tend that they have some power of reasoning. By reason
in this sense, we indicate that in every man by which he is
necessitated by his constitution, and as the condition of
his being a man, to have certain ideas and beliefs, so that
there is in every man a certain amount of mental furniture
that is common to all. These products of reason have received different names. They have been called "first
truths," "elements of human reason," "laws of belief,"
"principles of common sense," but in all the same thing is
meant. Perhaps a better name than either, as applicable
both to ideas and beliefs, would be rational intuitions.
These ideas and beliefs are not innate, but the capacity
for them is, and in such a way that they will infallibly
appear in every human being when the occasion for them
shall be given. The ideas are given at once, but the beliefs and judgments are not at first given in their general
form, but immediately assume that form through a particular instance, and not from a process of generalization.
They have a history and an order partly natural, and
partly in accordance with the history of the individual and
the process of his development. To say nothing of comparison, and of the ego and non-ego given in contrast, as
stated by Hamilton, the first act of consciousness must
involve, first, the idea of being; second, of action, since
consciousness is an activity; and, third, of the results of
the activity as of a thought or a feeling. Each of these
must, in the order of nature, be given before the idea of
personal identity, and this must be before that of causation. In regard to these we judge others by ourselves,
160
GRADATION OF INTUITIONS.
and have a right to. We believe that every event must
have a cause; we know that others believe it also, and
have a right to treat them as if they did, even if they
should deny it.
The ideas and beliefs which come to us thus may
be divided into first, mathematical ideas and axioms.
These are at the foundation of the abstract sciences, having for their subject quantity. In the second division
are those which pertain to mere being and its relations.
Upon these rest all sciences pertaining to actual being and
its relations. The third division comprises those which
pertain to beauty. These are at the foundation of asthetical science. In the fourth division are those which pertain
to morals and religion. Of these the pervading element is
the sense of obligation or duty. Of this the idea necessarily arises in connection with the choice by a rational
being of a supreme end, and with the performance of
actions supposed to bear upon that.
Here, again, as formerly, we find gradation. According
to the principles then laid down, abstract science is lower
than that of being; that of being, considered simply, is
lower than that of being in its beauty; and this is lower
than that of being self-directed and seeking its end under
a sense of duty. In the science of being abstract science
is implied; in the science of beauty that of being, and the
highest beauty is possible only in connection with duty
done. Here each higher implies the lower, but not the
reverse.
Of ideas and beliefs thus given, those that are moral are
so peculiar that philosophers have properly attributed their
origin to what they have called the moral or spiritual reason. This is reason, and something more, else it would
not be moral. This something more comes from its com 14*
161
LECTURES ON MORAL SCIENCE.
plexity as higher than mere primitive cognition or rational
intuition. In the first two classes of intellectual products
above mentioned, the intellectual element is almost sole.
In the third there is the synthesis of a rational product
with that of sensibility. But here, not only intellect and
feeling are involved, as in the mere contemplation of the
beautiful, the will is also reached. The idea of obligation
is nothing except as there is in it not only feeling, but a
requisition upon the will. As a product of the moral
reason is an idea, there is in it intellect; as it is an idea of
obligation, there is in it feeling; and as this feeling is that
of an imperative upon the will, it is clear that in a normal
state the activity of the moral reason would involve that
of the whole man. It is as nearly a synthesis of intellect,
feeling, and will, as is possible, and leave the will free.
Between ideas of the moral reason and others there is the
same difference as between a cannon-ball that is heated
and one that is not. They do not lie still and cold, but
respect action, and are of such a nature that we cannot be
indifferent to them.
This coalescence of ideas and affections, this fusion and
blending of them so that it is possible to give them but a
single name, together with their immediate proximity to
the will, is a characteristic of the moral nature that has not
been sufficiently noticed. In it we have moral ideas and
moral affections interpenetrating and moulding each other,
and thus a combination, as of light and heat, that is the
highest possible. As the product of the moral reason,
these ideas and the accompanying feelings arise necessarily
in all men; if they did not, we should not have a moral
nature; and because the moral reason is reason and something more, it raises us, according to our principle of classi
fication, to the highest grade of earthly, and, indeed, of
162
WILL.
conceivable existence. It will thus be seen, too, that moral
science must find its basis, not in any considerations of
outward utility, or perception of external relations, but in
the deepest and most fundamental intuitions of our nature.
The power by which moral ideas are thus originated,
originated by necessity, so that they spring up from the
depths of our being, together with the emotion that ac companies and forms a part of them, is an essential ele ment of personality. It is not something which the person
may use, but which being withdrawn, personality would
remain; it enters into its very framework. This is that
by which we are especially made in the image of God.
It is the organ of rational and spiritual intuitions. It is
not exhausted by those ideas which all men must have that
they may be men, but being held in right relations, it is
capable of receiving, and in the progress of the man is
necessitated to receive, new and higher ideas, still having
the same characteristics of universality and necessity for
all who reach the same point.
We saw in the third lecture how man is connected with
all that is below him, through the laws that govern all
below, and extend up to him. We now reach the point at
which he is, or has the capacity to be, connected with that
* which is above him. As rational and in the image of God
he must have, in kind, the capacities of the very highest
creature, and be subject to every fundamental law of the
spiritual world. The laws of that world reach down to
/ him, as those of the world below reach up.
After reason the next element of personality and con dition of moral action is a Rational Will, - a Will in Free dom.
Without freedom of some kind, connected with an act
at some point, all are agreed that there can be no obliga
163
LECTURES ON MORAL SCIENCE.
tion or responsibility. A man is not responsible for the
movement of the earth in its orbit, because it has no connection of any kind with his will. To awaken a sense of
obligation in regard to anything which has thus no connection with the will, direct or indirect, proximate or
remote, is impossible. When, therefore, we see a man perform an act that we call moral, the element of will and of
choice is presupposed.
By some, by most indeed, this element of will is sup posed to be the chief one in personality, and there are
those who regard it as the only one. Others again think
of it as the executive of a person already constituted. To
me it seems that the moral ideas that are given by reason,
in the light of which we choose and act, through which,
indeed, the will is a rational instead of a brute will, are
quite as necessary to personality as the power of choosing
and acting, and that both are indispensable.
But with these two, - reason and free will, including
moral ideas and affections, and so conscience, - whatever
we may think of the part that belongs to each, we have
the apriori elements of personality, and so the power of
doing a moral act.
I have spoken of personality as composed of elements.
It seemed necessary to speak thus; and yet I am inclined
to think that our idea of a person is simple. A person is
something more than reason and will. We get misty and
lose ourselves by always using abstract terms and the
names of attributes. A person is a substance, a being that
has reason and will. Here we reach an agent, and the
true point of responsibility,- the man himself. It is the
man himself, the person, the self, the ego, the me, whatever
you please to term it, that we hold responsible and praise
or blame. It is this mysterious -- mysterious as all things
164
A
CONDITIONS OF A MORAL ACT.
are that are simple - this mysterious and inscrutable per son, this self-conscious, thinking, comprehending, electing
being,- it is the man himself that we approve or disap prove. Aside from their origin in him, actions have no
moral quality. Constitutional tendencies, desires, affec tions, have no moral character till he adopts them, and
' consents, or elects that they shall move in a particular
direction.
Of a person thus constituted the three characteristics
are that he is rational, free, and moral. Such a being may
perform acts merely instinctive; but as the moral reason,
with its necessary products of moral ideas and affections,
enters as an element into the conception of personality, it
can never be optional with him whether he will have a
moral character. He must have one involving the very
essence of his being, and his only option is whether it shall
be good or bad.
We have now the powers prerequisite to a moral act.
But there is another condition. A moral act must be
also rational, and as such must have reference to an end.
This necessity which a rational being is under of acting
with reference to an end, so that his doing this is a test of
his rationality, would seem to imply that his conception of
an end is the fundamental one for man as an active being.
As has been said, the ideas of reason have a history and an
order. For man, as speculative, the idea of existence is
first. It is implied in all assertions respecting identity
and causation. In the same way, the idea of an end in volving a good is implied in all acts of rational choice.
As a will, rational and free, is essential to morality, so must
everything be that is a prerequisite of the action of such
a will. But to a free and rational act of willing, the con- S
ception of an end is necessary. The moral sentiment, or
165
LECTURES ON MORAL SCIENCE.
conscience, is evolved only in connection with the action
of free will with reference to an end. This may, therefore,
be considered as the fundamental and primitive conception
of man as active and moral.
But if reason would act reasonably it must not only
know an end, but its own end. Would, then, a rational
being naturally know his own end, or what he ought to
choose? If not, he would be lost. Without a capacity
for this, such a being would be an absurdity. The gropings of a baffled instinct would be nothing in comparison with his blindness and helplessness. This end may be
known, either from the insight of reason, or firom revelation; but however known, there must be a capacity in
reason to recognize its own end; and the test of such an
end as adequate must be that it shall always suffice to
call forth the highest normal activity of the highest powers.
In anything short of this a want would be felt. In such an
end, if we consider the capacities, the worth and grandeur
of spiritual being, there may be an infinite good. There
may and must be that which should cause it to be adopted
by the whole energy of the will.
We have now the prerequisites for a moral act. We
have a person knowing his end. But a rational being
knowing his end cannot but know his law, since the law is
revealed in the end. Thus, and thus only, can he become
"a law to himself." It is in the apprehension by a person
of his end that the moral nature manifests itself in the
immediate and necessary affirmation of obligation to
choose that end. This is the moral law, and the whole of
it. It is wholly spiritual, simply requiring choice. Lying
in immediate proximity to the will, it cannot become a
rule; no means can be used; and nothing but a want of
will can prevent its being obeyed. Obligation to choose
166
HOLINESS.
the end is affirmed in view of it as good; and such a choice
is approved as right.
It is here that we find the point of coalescence between
intuitional or a priori systems of morality, and those that
are inductive. At some point these must come together,
for it is impossible that the great thinkers in either line
should be wholly wrong. The intuitional element here
finds its sphere in the immediate recognition by reason of
its own end, and in the necessary affirmation of obligation
to choose that end. The practical nature, asking, "Who
will show me any good?" is also satisfied, because the end
thus chosen is a good, and the good; and because there is
in all questions of right a constant call for the activity of
the inductive powers.
By some beings it may be that their true end alone is
seen and embraced. They may know no other as possible,
and so never be tempted. But for others there is an
alternative so presented that there must be a choice between this end and its opposite. Let now the true end be
chosen, and the star finds its orbit; there is moral order,
there is peace, there is "joy in heaven."
Choosing thus his end, with an apprehension of the worth
of spiritual being, with a consciousness of worthiness in
having thus chosen, such a being would move on in peace,
-not the peace of quiescence, but of a tranquil and deep
joy, -till there should arise from within or from without
some disturbing influence that might come between him
and his end. Then, in proportion to his sense of the
worth of the end, and of the obligation to choose and
seek it, must be his abhorrence and condemnation of an
opposite choice, and his opposition to anything that would
divorce him from his end. HIence virtue is necessarily
bi-polar. As such, it becomes holiness. This is reason
167
LECTURES ON MORAL SCIENCE.
vindicating its right to attain its end. It is personality
expressing its sense of the value of its end, now in complacency with it, and with all that would promote it, and
now in indignation and opposition towards all that would
oppose it. From the evolution on the one side comes
all that is mild and winning in virtue; from that on the
other all that is stern and awful. It is by the term holiness - that is, wholeness- that this double aspect, and so
the completeness of virtue, is best expressed.
That the above may not seem opposed to our consciousness, it may be well to state that in choosing a supreme
end it is not necessary that we should know or choose it
abstractly and formally, but simply that our individual
and specific choices should involve it, and be instances
under it. So it is that we know and act under the idea or
principle of causation, and so under mathematical axioms.
The act of a child may involve the axiom that the whole
is equal to the sum of all its parts, and yet the child may
never have heard of the axiom, and in that form could not
comprehend it.
In thus choosing a supreme end, if that end be the good
of others, we reach the highest significance of the word
love. This is an act both of the affections and the will,
and carries every faculty and choice of the soul along with
it. In it the man disposes of himself. It lies back of
specific choices and volitions, and determines character.
Springing from a synthesis of the rational sensibility and
the will, it is the highest product of our highest powers,
-the consummate flower of our existence.
From what has been said above, we shall readily see
what that form of activity is to which responsibility ultimately attaches. It is not volition regarded simply as an
executive act; it is preference. It is that immanent act of
168
THE POINT OF RESPONSIBILITY.
preference in which we dispose of ourselves, and on which
character depends. It is this that gives its set to the current of the soul, and determines the character of subsequent specific acts of preference and volition under it. It
is an act of will, as distinguished from the feelings. It is
either that impartial love which is commanded by the
moral law; or a giving up of the soul to be governed by
the propensities. It is at this point that we find moral
freedom. That the ultimate point of responsibility must
rest here, appears from the effect of such a preference in
controlling the thoughts and modifying the feelings, and,
as thought and feeling act and react upon each other, in
changing the very principles of association. Nothing is so
cunning of fence as such an underlying preference when
anything would interfere with it. As already intimated,
it may so control the laws of evidence as perceived by
us, that a man shall really believe and act upon a lie, and
mistake the reality of such belief for that genuine sincerity and coming to the light of which our Saviour speaks.
Hence it is that a man may verily think that he is doing
God service while he is persecuting his people, and doing
his utmost to overthrow his cause in the earth.
The word "intention" is often used by moralists to indicate what is ultimate when they would reach the source
of morality; but it does not do this. Intention refers to
specific volition, and implies an opportunity, real or supposed, to carry out the intention. Hatred of a person
whom we were sure we never could reach would not be an
intention, nor would it give rise to any intention of injuring him. All intentions that indicate character spring
from some form of settled preference, which may multiply
itself in such intentions without number or exhaustion.
15
169
LECTURES ON MORAL SCIENCE.
Hence this preference, which in the Scriptures is called
the heart, is compared to a fountain.
At this point there seems to be a general agreement
among writers on morals on three things: The first is, that man is responsible for his preferences,
his choices, the acts of his will generally, -for these and
their results, - and for nothing else. It will be found that
those writers, as Edwards, who speak of man as responsible for the affections or heart, either regard these as
synonymous with will, or as a part of it. Says Edwards,
"The will and the affections of the soul are not two faculties; the affections are not essentially distinct from the
will." There are, indeed, some whose language might lead
us to suppose that they hold to an inherent moral quality
in affections that are purely spontaneous; but on reflection
it will be found impossible to attach responsibility to a
being incapable of rational preference, and so of the choice
of an end.
It is agreed, in the second place, that there is a broad
distinction between what is called, sometimes an immanent preference, sometimes a governing purpose, sometimes an ultimate intention, and those volitions which are
merely executive, and precede specific acts under such a
purpose.
In the third place, it is agreed that character is as the
governing preference or purpose - that it consists in an
original and thorough determination by a man of himself
with reference to some end chosen by him as supreme.
In connection with the choice of a supreme end all the
phenomena of a moral life are evolved. In view of the
end there arises, as has been said, a sense of obligation to
choose it. From these two arise the idea of moral law;
for moral law is the affirmation by reason of the obliga
170
MERIT AND DEMERIT.
tion to act rationally. A divine law is the same law proclaimed by the authority of the Infinite Reason, and acconpanied by sanctions. So is it that while the man is a law
to himself, the divine law is recognized at once as the law
of the inner life; and so will its full revelation, if the inner
law has become obscured, be but as the clear light of day
after the dim twilight. It will not be a thing wholly new
and strange, but homogeneous, and but the increase and
fulness of "that light that lighteth every man that comethli
into the world." It is into this light that men may dome
more fully and walk in purity, or they may withdraw from
it, and walk in darkness.
After the ideas of obligation and of law, must arise those
of merit or demerit, of self-approbation or of self-condeinnation, as the true end, or its alternative, has been chosen.
Merit and demerit are supposed to arise chiefly in connection with something done outwardly, but if the end be
chosen with a paramount affection, as a supreme end must
be, outward acts according with the choice will follow of
course. These simply indicate the strength of the inward
principle, and in that is the only merit.
Again, in a sense of merit or demerit there is not only
a present satisfaction or dissatisfaction, but a promise or a
threat for the future, and these may become elements of
great power. We thus get the notion of reward and punishment, and through these of responsibility, for, if there
were no reward and no punishment, there could be no responsibility. It is at this point that the moral nature of
man is connected with the government of God as outwardly revealed. If there were no consequences of acts
in the way of rewards and punishments through the will
of another we could not be under the government of that
other, or responsible to him; and if those consequences
171
LECTURES ON MORAL SCIENCE.
should have no reference to merit or demerit, the govern
ment could not be moral.
Thus do we find, in immediate connection with the
choice of a supreme end, the ideas of obligation and of
moral law, of merit and demerit, of reward and punishment, and of responsibility. We find also the ideas of
right and wrong. Properly these are always relative, expressing either fitness or unfitness, and having reference to
an end. As such they are secondary; but they imply a
moraI quality when they indicate the fitness or unfitness
of specific moral acts, or of the fundamental position of
the heart with reference to the true and supreme end.
As the ideas and feelings just mentioned arise in our
minds, a tribunal will be erected within us by which we
shall be compelled to judge ourselves, and by which we
shall also judge others in accordance with what we suppose to be the character of their radical choices. Without
such a tribunal, and power and necessity of judgment, our
moral nature would not be complete. There would be no
answering of face to face, and we should not be linked in
sympathy with the one great community of moral beings.
As illustrating the gradation and classification of ideas
heretofore referred to, it may be well to say, at this point,
that the highest forms and ideas of beauty and sublimity
are also evolved as subsidiary, in connection with the
choice of a supreme end and its results. In all working of
unconscious and involuntary powers towards their end, and
the facile mastery by them of the material to be used and
the obstacles to be overcome, there is beauty. Virtue is
the same thing when the powers are conscious and voluntary. Hence their deep affinity. There is no beauty of a
ship with every sail set, speeding its way over the subject
element to its haven, that can be compared with that of
172
CONSCIENCE.
the organized powers of man acting in harmony, - those
ruling that ought to rule, and those serving that ought to
serve, and all conspiring to their destined end; nor is any
storm in nature so sublime as the conflicts that may arise
when temptation and opposition come between a truehearted man and the attainment of his end.
It is somewhere in connection with the central act of
choice now spoken of that conscience must be found.
Of the discrepancy there is in the views respecting conscience, I spoke in the first lecture. This discrepancy cannot be removed at once, if at all. It arises from the
intimate blending there is in this higher nature of the
powers of knowing and of feeling, so that we may and
do call the product indifferently an idea or a feeling.
Thus we say, the idea of obligation, and the feeling of
obligation. Hience some have regarded, and probably will
continue to regard, conscience as comprising the whole
moral nature. "The moral nature of man," says Dr. Alexander; of Edinburgh, in a learned article recently published in the Encyclopedia Britannica, -" the moral nature of man is summed up in the word conscience. Moral
nature and conscience are two names of the same thing.
The analysis of conscience, therefore, will unfold man's
moral nature." I prefer a view which makes the operation of our moral nature more analogous to those of the
other departments of our complex being. In all of them
there was original provision for the right performance of
their work; for a recognition of the character of that performance as normal or otherwise; and in that recognition
for a sense of satisfaction or the reverse, which may be
regarded as reward or punishment. So with the moral
nature. It was intended that it, or rather the person,
should work in accordance with his law. If he does so,
15*
173
LECTURES ON MORAL SCIENCE.
there is in it a testifying state that is not only recognition,
but approbation and reward. If he does not, there is also
a testifying state that is disapprobation and punishment.
Conscience, then, will involve a recognition by the person
of the moral quality of his own acts or states, and the
feelings consequent upon such recognition. It may be
defined as that function of the moral reason by which it
affirms obligation before the act, by which it approves or
disapproves after the act, and by which it indicates future
reward or punishment. Here, high as it is, we still see in
it an analogy to appetite. In that, as in hunger, there is
both impulse and discrimination, and there is subsequent
pleasure or the reverse. To the prophetic power of conscience, however, appetite has nothing analogous. Conscience will then reveal itself as, 1st. Obligatory. 2d.
Judicial. 3d. Prophetic. There will be, first, the affirmation of obligation before the act; second, the excusing or
accusing by one another of the thoughts after the act; and,
third, a promise or threat that becomes, on the one hand,
a hope of eternal life, or, on the other, "a certain fearful
looking for of judgment."
By many, by most, conscience is regarded as a separate
faculty, and, as has been said, the whole of the moral nature. I prefer to say, as above, that it is a function of the
moral reason. Besides affirming obligation to choose the
true supreme end, the moral reason is that in the light of
which it is chosen. It is that by which that end is recognized as supreme. The affirmation of obligation, as above
stated, is what many mean by the apprehension of an ultimate right.
On this subject writers generally begin by assuming that
there are actions having a moral quality, and regard the
conscience or moral nature as that by which we perceive
174
DOUBLE FUNCTION OF MORAL NATURE.
and become affected by that quality. But whence came the
moral act? From a moral being certainly; and we should
naturally suppose that those capacities by which a being
could originate an act having a moral quality would be
the leading part of his moral nature, rather than that by
which he should perceive and become affected by the
moral quality after it was originated. Is our moral nature
that only by which we approve and condemn? Or is it
that also by which we originate and do the things that we
approve and condemn? We love God. By an act of our
moral nature we approve ourselves in so doing. Is it by
an act of our moral nature that we love him? I suppose
it is. We do not love God because we are under obligation to, except as his worth and worthiness impose the obligation. We love him impartially because of his worth,
and complacently because of his worthiness; and such love
is from our moral nature, but not from conscience. If the
states or forms of activity judged did not have a moral
quality they could not be approved or condemned, and
they belong to our moral nature in virtue of their having
a moral quality. That also by which we judge belongs to
our moral nature because it judges of moral quality.
In the order of nature there must be a moral being before there can be a moral act. But, as we have seen, a
moral being is a person having moral reason and the moral
ideas and affections necessarily originated by that, together
with firee will, which is implied as a condition for the formation of those ideas. In these is personality and a moral
nature -the capacity of doing a moral act. But these are
not conscience. That becomes possible only when there is
a question respecting the conformity, future or past, of a
being, already moral, to what either is, or is supposed to
be his law.
175
LECTURES ON MORAL SCIENCE.
A moral act is one that respects the supreme end. Any
act which a man may either do or leave undone, and still
stand in precisely the same relation to his supreme end, is
not a moral act; and the moral nature will comprise, as
I have said, all that is ultimately directive - the moral
reason, the will, the personality, the man himself, that
which does the moral act, as well as that which judges
of it.
We may, then, regard the whole moral nature as consisting of those powers whose activity gives the moral
quality, and also of those which judge of the moral quality and are affected by it; and it would conduce to perspicuity if the term conscience could be confined to the
latter.
Of the moral quality itself which conscience presupposes, our notion is simple, as of color or extension. We
perceive it immediately as belonging to certain states of
mind, as selfishness, envy, malignity, on the one hand,
and benevolence, generosity, and kindness, on the other.
Relations may be needed to evolve the acts, but it is from
no perception of them. It is from no sense, but is an immediate knowledge, by the spirit, of the quality of its own
states and acts. We know a moral act as moral precisely
as we know an intellectual act as intellectual. We know
an intellectual act to be intellectual because it is an act of
the intellect; and what an act of the intellect is, and that
it is intellectual, every being having an intellect must
know intuitively on the exercise of his intellect, and he
could know it in no other way. Here is primitive knowledge, without which no definition could give the first elements of the knowledge of anything. It is in the same
way that a moral act presupposes a moral constitution, and
is known to have a moral quality.
176
CONSCIENCE PROPHETIC.
In the definition of conscience the prophetic element
requires special attention. It is important not only practically, but as proving the being of a personal God, and as
connecting us with his government. Not only would it
require a God with will and freedom, and an apprehension
of moral distinctions, to originate a creature endowed with
these, but without such a being as a rewarder and punisher, the idea of reward and punishment other than that
which is natural and immediately inflicted, is nugatory.
Without such a being it would be a practical absurdity -
an eye without light, a part in nature without its counterpart, a falsehood in the very sanctuary of the moral nature
of man.
In strictness I suppose the office of conscience to be to
take cognizance of our own moral acts, and that a decision
respecting those of others should be referred to the judgment. That this is its proper sphere appears from the fact
that conscience was not originally, in our own language,
and is not now in some others, distinguished from consciousness. It was consciousness par eminence; but on a
subject like this, whatever may be thought of Hamilton's
view of it, consciousness can respect only what passes
within ourselves. In the common definitions and descriptions of conscience, powers are assigned it, as that of impulsion, and of rewarding and punishing, which must have
reference solely to our own acts. This, too, would seem to
be the scriptural idea. Paul says of the Gentiles, " Which
show the work of the law written in their hearts," - that
is, in their moral nature, - "their conscience also bearing
witness, and their thoughts meanwhile accusing or else
excusing one another." In all this there must have been
reference solely to their own acts.
In thus finding a moral nature, and so a person, with the
177
LECTURES ON MORAL SCIENCE.
power both of doing moral acts and of judging of them,
we reach the highest form of created being. It is, doubtless, the highest possible, since there is in it the image of
God, who is himself a person. We reach that for which
all else is a condition, and which has, therefore, over all
else, as below it, a natural supremacy. By a natural law
"all sheep and oxen, yea, and the beasts of the field, and
the fowl of the air, and the fish of the sea," are put " under
the feet" of such a being. In the same way everything in
his own system that is physical, or animal, or merely intellectual, must either be in subjection, or in disorder and
rebellion. We now reach a form of activity that is a condition for nothing within the system, above itself, which
has in itself and in its results not only a good, but the
good, and the supreme good for man, and which can,
therefore, be subject to no law of limitation.
In the exigencies of the present life it may happen that
there shall be not only limitations, but exceptions to the
laws of every subordinate portion of the system. The
laws on which the welfare of the body depends may be
disregarded, because the welfare of the body is not the
highest good. There may be virtue, and even heroism, in
disregarding them. But the laws of the moral nature cannot be thus disregarded. Than the end and good which
these laws would secure, there can be nothing higher;
there can therefore be no law to which these can give
place; they can be subordinate to nothing, are always
binding, have no exceptions, and the activity under them
can never be in excess.
Having thus reached what is highest in man, we must, in
accordance with our previous discussions, here find his true
end and good. And here we do find it in the activity of
the personality according to its law. What then is that
178
NATURAL AND REVEALED LAW IDENTICAL.
law? The law of the subordinate faculties is, an activity
for each upon its appropriate object up to the point aJ
which it would interfere with some higher form of activity
and good. Now, however, there can be no interference
with anything higher. The law, therefore, of the highest
faculties will be their highest possible form and degree of
i activity upon their appropriate objects. What, then, is the
highest form of activity of which we are capable? By a
fair analysis this has been shown to be love. What are the
appropriate objects of love? They are God and our neigh bor. What is the highest possible degree of this love? It
is the love of God with all the heart, and of our neigh bor as ourselves.
Here, then, do we have, after as full and fair an examin ation as I could give it, the human constitution itself utter ing the substance of that law which was spoken in thunder
four thousand years ago, and uttering, because it is impos sible to find those more appropriate, the very words of
Him who spoke as never man spoke, when he gave a sum mary of that law. Wonderful is it that his words should
be the exact formula for the expression of the highest pos sible activity of the highest powers.
Thus, as in a former lecture we found that the teachings
inwrought into the whole frame-work of nature were in
perfect harmony with the constitution of man, so do we
now find that the teachings of that constitution are them selves in perfect harmony with those of the revealed word
of God. So is it that "deep calleth unto deep." So is
man the connecting link between that which is lowest and
that which is highest.
We have now answered the three questions put in the
second lecture. The first was, What ought man to do?
The answer was, To choose and seek the end for which
*
179
LECTURES ON MORAL, SCIENCE.
God made him. The second was, Why ought he to do it?
The answer was, Because of the intrinsic good there is in
the end. The third was, tIow ought he to do it? The
answer is, By the highest activity of his lower powers
according to the law of limitation; and by the full activity
of his highest powers upon their appropriate objects.
Does any one inquire more especially what this activity is?
The answer is, Since we have shown the moral affections
to be higher than the intellect, and since God is the highest and only adequate object of the affections, that it can
consist only in the supreme love of God, and the impartial
love of man.
180
LECTURE VIII.
RELATION OF VIRTUE TO HAPPINESS.- QUANTITY AND QUALITY OF GOOD.
- MORAL AND NATURAL GOOD.- REGARD FOR OUR OWN GOOD. - CON NECTION WITH BENEVOLENCE.- ENJOYMENT FROM APPROBATION. THE TRUE END OF MAN.-CONNECTION BETWEEN MORAL AND NATURAL
GOOD.
THE identity which we found in the last lecture between
the teaching of the constitution of man and the law of
God was not sought. The result was reached because the
analysis would go there. I was myself surprised at the
exactness of the coincidence. The formula we reached
for the end and good of man was the highest possible
activity of the highest powers upon their appropriate
objects. Love has been shown to be the highest form of
activity; and how readily and perfectly the law of God
takes the form of the above expression will be seen if we
observe that no love of him can be greater than that with
all the heart, and no love of our neighbor can be greater
than that we should love him as ourselves.
It is a grand and beautiful thing thus to begin, as we
have done, at the foundation of this lower creation, and to
follow it upward as its stories rise one upon another till
they culminate in man, and then to hear from his constitution an articulate utterance identical with an utterance
from heaven that comes down to meet it. So is man fitted
to be a being, as Milton says, -
"Commercing with the skies."
The teachings of the constitution, or of natural law, being
16 181~~~~~~~~~~~~~~~~I
16
181
LECTURES ON MORAL SCIENCE.
thus identified with those of the revealed law, it would now
be in order to go on and evolve the specific duties that
would flow from this law as applied in the various relations of life. This might be done, as it generally has been,
in the light of that disposition which would lead us to do
good to all men; or, more properly, as more in accordance
with the preceding course of thought, in the light of
ends. The duties of man to himself and to God would
then be determined ill the light of his end as a creature of
God; his duties in the family in its various relations would
be determined by the end of the family, and his duties to
society by the end of society. And this it was my purpose at one time to do; but that would be beaten ground;
the time would not be adequate, and there are still speculative questions of interest, that are also practical, that
require our attention. We need particularly just now to
analyze this love with reference to certain general conceptions that have been formed, and their harmony with each
other. We need to inquire after the relations to each
other of holiness or virtue, and happiness.
The revealed law is practical. It applies its precepts
directly to a person; it says thou; and it requires duties
to be performed towards persons. The objects are God
and our neighbor. But the mind forms necessarily certain
general conceptions. These are represented by general
terms having no reality or one thing in nature corresponding to them, but simply the notion as it is formed in different minds, and which may vary much, both in its content and in its distinctness. The general notion of property may be in some minds clear, in others indistinct; in
some it may be represented by land, in others by stocks.
These general terms, formed by abstraction, and thus varying in their significance in different minds, have been
182
HOLINESS AND HAPPINESS.
thrown into the arena of discussion, and bandied about
endlessly. So will they continue to be; the terms, as we
may hope and believe, becoming in the mean time more
definite, the conceptions of men in connection with them
more distinct, and their relations with each other better
established.
Such conceptions, those too which have been the subject
of much discussion, we shall find involved in the general
formula already reached. Those to which I refer are the
conceptions of holiness and of happiness. What is needed
is that these should be uniform and distinct in the minds
of men, and that their relations to each other should be
clearly seen. There is a natural feeling that virtue, or
holiness, and happiness ought to be united. Moral order
seems to require this. In this world they appear to be
often separated, and hence the strangeness of that state in
which this world is.
That holiness and happiness can be identified as objects
of pursuit is denied by Kant, and it is in their separation
that he finds what he calls the "antinomy" of the practical reason. According to him "the connection between
them is not causal." "Man is bound to pursue virtue;
man cannot but pursue happiness; and yet neither are
these identical, nor does the one lead to the other." Of
old the doctrine of the Epicureans was that "to be consciously influenced by maxims that lead to happiness, is
virtue." The doctrine of the Stoics, and the opposite of
this, was, that "to be conscious of virtue is happiness."
"The identification of happiness with duty," says Whewell, "on merely philosophical grounds, is a question of
great difficulty." Possibly our past discussions may throw
some light on this point.
In estimating enjoyment or good, regard must be had to
183
LECTURES ON MORAL SCIENCE.
both quantity and quality. The quantity from any given
susceptibility or power will be as its normal activity. The
quality will be as the rank, according to the gradation heretofore indicated, of the susceptibility or power. There are,
I know, those who say that the only difference in respect
to enjoyment is in degree. So Paley thought. They say
that the enjoyment of the glutton is just as excellent and
valuable as that of the saint or angel. Do you believe
this? Do you think that any amount of swinish enjoyment could be weighed against one hour of the clear comprehension of God and his works, and of sinless and fervent love? I greatly mistake if there be not in the common consciousness of men, as there is expressed in their
language, a feeling of gradation in respect to enjoyments
that corresponds substantially with the order of the faculties as heretofore explained. When, however, we come to
the moral nature, as we there make a leap in respect to
the order of the faculties, so do we in respect to the kind
of enjoyment. As we now come to have faculties like
those of the angels, and are made in the image of God, so
do we become capable of enjoyments like those of the
angels and of God. Between such enjoyment and that of
an animal, or of our own animal nature, there is as much
difference in dignity and worth as there is between an
angel and an animal. Here only do we find moral and
spiritual enjoyments; here approbation and disapprobation; here the consciousness of worth.
The above being premised, we say that the natural law
and formula for the highest enjoymnent is the highest possible activity of the highest powers upon their appropriate
objects. We say, also, that the formula for virtue is the
highest normal activity of the moral powers. But thte
moral powers are also the highest powers, and hence the
184
MORAL AND NATURAL GOOD.
highest enjoyment must be in and firom the same activity
in which virtue consists. If, then, they mnay not be said to
be identical, they are inseparably connected by a Tatural
law, as much so as the light is with the sun. It is one of
the properties and characteristics of the sun by which we
define it, and as God made it, that it gives forth light; and
it is one of the properties and characteristics of virtue as
we always conceive of it, and as God intended it should
be, that it gives forth its own natural, inseparable, peculiar
enjoyment. It is an enjoyment that belongs to it, and
inheres in it, as the property in its substance; so that the
Stoics were right in saying that "the consciousness of
virtue is happiness."
This brings us to the distinction between what may
be called moral good and that which is merely natural.
Moral good is that which is immediately, and by a natural
law, connected with the normal activity of the moral powers; natural good is that which comes from the activity of
any of the susceptibilities or powers below those that are
moral. They are alike in being instances under the general law that there is from the activity of each faculty its
own enjoyment; and in that sense both are natural; but
what I have called moral good is not only the product
of the moral powers, - it has peculiarities well worthy of
notice, and such as to fit it to be the good of the race.
One of these is its independence. By this I mean that
it is wholly within the control of the man himself. This
arises from the fact that moral good is from the direct
activity of the will itself, and not fron the activity of
those faculties that depend on the will. Tyranny may fetter the limbs; want of discipline may render the faculties
indocile; but virtue consists in the voluntary acts themselves, and in those voluntary dispositions which lead to
16*
185
LECTURES ON MORAL SCIENCE.
the acts, that is, in the activity of the will. This is central
to the man. It is the man himself acting, and nothing can
come between him and it, together with its natural results.
It is not these results that are meant when we say we will
do right and leave the result with God. These we conceive of as included in doing right. It may even be
doubted whether, moral beings existing, the results could
have been otherwise. The dispositions and volitions are
one thing, the command of the faculties through which
these express themselves is another. In the one is character; in the other ability. Any object of our desires we
may be prevented by external circumstances from obtaining; but no will of another, no violence or imprisonment,
no external circumstance can come between a man and his
voluntary dispositions, together with the blessedness there
may be from their activity.
This puts the highest interest of every man into his
own power. If he have confidence in God, it gives him
a rational ground on which he can stand and be a martyr.
Here is a citadel that can never be forced; if it surrender,
the man himself must open the gates. In respect to this,
the exhortation may be fairly given as against any external
influence, " Hold that fast which thou hast, that no man
take thy crown." It is in this power of man thus to resist,
in his allegiance to virtue and to God, all solicitation and
all violence, that his true greatness is found.
It is at this point, as the will is differently related to the
grounds of its action, that moral beauty and moral sublimity arise. When the propensities and faculties yield themselves in ready and glad coincidence with the virtuous
will, when other moral agents conspire with it, and nature
is accordant, there is moral beauty. There is no temptation then, and the current of the soul flows on without a
186
MORAL GOOD.
ripple. But when the propensities and faculties are refractory, when they solicit to evil, and would fain rebel; when
example and authority are against us so that integrity
would require resistance unto death, then, if the will remain firm, there is moral sublimity. In the one case the
element is spontaneity, consent, and harmony of action;
in the other it is force, struggle, victory. In both there is
a sense of dignity, of freedom, and self-direction. There
is the joy of the young eagle when he poises himself on
his own pinions, and that something more which the eagle
cannot feel that is involved in self-approbation and a consciousness of merit.
This leads us to a second peculiarity of moral good. It
is that it is necessarily accompanied by a sense of approbation. This is an element wholly unknown till we reach
the action of the moral powers. Up to this point we have
a pleasure in all excellence; we admire it; but when we
reach moral excellence, admiration becomes approbation.
This gives a pleasure entirely distinct from that naturally
connected with moral goodness. In the love of God or of
man there is an enjoyment wholly distinct from the approbation. That is in view of the love, and subsequent to it.
Love and hatred have in them respectively the elements
of happiness and of misery, aside from any subsequent act
of approbation or disapprobation. It is in these subsequent acts that we find a consciousness by the spirit of its
own state as it is, or is not conformed to the law of its
being, involving a feeling of self-approbation and hope, or
of self-condemnation and of an indefinite dread. As virtue is in the states and acts of the will, so, if there be
candor, the eye of conscience is directly fixed upon these
states and acts; and so distinct at times do these sentiments of approval and condemnation become that they
18T
LECTURES ON MORAL SCIENCE.
seem like the product of a second personality within us,
recording our deeds, passing sentence upon them, and giving dim forebodings of a coming and more perfect judgment. This is, indeed, the great, and almost the only evidence from nature of a future retribution; at least, it is
that without which no other would have any weight.
liere, then, in connection with moral good, we have a
new and striking element, and one which we should suppose could hardly fail to direct men to that as their chief
good. This approbation is not, as is sometimes supposed,
the good itself, but comes in as the accompaniment, the
sanction and heightener of that good.
A third peculiarity of moral good is that in seeking it
for ourselves we necessarily promote the good of others.
We thus find a coalescence of what is called self-love with
benevolence, and of interest with duty. In this perfect
coalescence and harmony is the point of reconciliation between what have been called the selfish and the benevolent
systems of morals. By some it has been held that all virtue has its origin in a regard for our own good; by some
that it consists in a regard to the good of others. The
true system is found in the coincidence of the two; and
that becomes possible only from the peculiarity of moral
good now mentioned.
This point requires attention, not only because different
and seemingly opposite systems have sprung from it, but
also because there has been in the public mind, to some extent, a wrong estimate of what has been called self-love, or
rather of the right and the duty of every man to seek his
own highest good. As indicating this right and duty, selflove has not been a fortunate term. It has not always
been clearly distinguished from selfishness, and, if not positively wrong, has been supposed to be less noble and
188
INDIVIDUAL GOOD.
worthy than benevolence. Both the element of duty that
is in it, which ennobles all things, and that of beneficence,
which is also in it, have been overlooked. To exhort men
to love themselves has been supposed to be a work of
supererogation, if not positively wrong. We need, therefore, to say a word on this point, and then to show how
the two coalesce.
If the terms are rightly understood, we need not hesitate in saying that a man cannot love himself too much.
Does this startle any one who has been accustomed to a
particular form of phraseology? I would ask him whether
he thinks we can love others too much? If not, neither
can we ourselves, since the love of ourselves is made in
Scripture, as it must be by reason, the measure of our love
to others. "Thou shalt love thy neighbor as thyself."
Let a man then love himself as much as he will, only let
himlove his neighbor as much. Let him love his neighbor as much as he will, only let him love himself as much.
This is the Bible doctrine, and in this equal and impartial
love the good of the whole will be provided for, since both
the individual and his neighbor are equal factors in making up the great sum of good.
Again, that it is the duty of the individual to seek his
own highest good is involved in his structure. He would
be a reproach to his Maker if it were not. It has already
been seen to be the characteristic of a rational being to
act with reference to an end. But an end can be sought
rationally only as there is in it an apprehended good. It
would seem, therefore, that the idea of good must be among
our primitive and elementary ideas, as much so at least as
that of an end. It will probably be found to result at once,
immediately, and always, from any normal activity of the
powers. Without a supreme good, man would be a coin
189
LECTURES ON MORAL SCIENCE.
tradiction and an absurdity. If there were no good for
man he could do nothing for himself, and nothing could
be done for him. Without a rational conception of good
there could be no rational activity; and that which is thus
at the basis of all action of the rational powers we may
well suppose to be primitive. In the analysis by Cousin
of our moral ideas, he says that those of merit and demerit,
of justice and injustice, of right and wrong, must precede
those of reward and punishment. This is true, but it does
not follow that they must precede the idea of good in
general. The idea of good certainly underlies those of
reward and punishment; but since justice must consist in
bestowing good and inflicting evil, it would seem that it
must underlie that also, and that there can no more be a
conception of justice and injustice in general without that
of some end or good, than there can be one of commercial
justice and injustice without that of an exchangeable
value. But that the ideas of justice and injustice are natural and necessary, no one doubts.
In connection with this idea of good there must be
some tendency towards it, or there could be no harmony
in the being himself, and we could have no conception of
him as acting morally. Towards the supreme good there
must have been some constitutional impulse, as well as
towards those that are lower, since it must have been
intended for a motive, and for that mere comprehension
would not suffice. It may indeed be doubted whether
good could be conceived of as good, without such a tendency. This tendency may have become enfeebled, obscured, confused; but no philosophy of ends could be
conceived of without it, and unless the nature be hopelessly ruined something of it must remain. For a nature
190
TENDENCY TO INDIVIDUAL GOOD.
thus ruined, all attempts at a philosophy of itself would
be simple bewilderment.
This tendency is always implied both in speculation and
in action. By many it has been called a rational instinct.
Archbishop Leighton went so far as to use for it the term
"appetite." "Actual or formal felicity," says he, "is the
full possession and enjoyment of that complete and chief
good (that, namely, which most perfectly supplies all the
wants and satisfies all the cravings of our rational appetites)." Of this tendency McLaurin says, "God has implanted in us that thirst after complete happiness which is
the spring of men's actions; and since the above-mentioned
faculty of reason shows where that thirst may be satisfied,
the direct tendency of both, if duly approved, would be
to lead the soul to the eternal fountain of all good." It
was of this that the schoolmen we?e wont to say, "In
beatitudinem fertur voluntas, non ut voluntas, sedl ut
natura" -"The will is borne towards happiness, not as
will, but as nature." This is what was meant by the
psalmist king, in whom this tendency worked in the light,
when he said, "My soul thirsteth for God, for the living
God." "My heart and my flesh crieth out for the living
God." This it is which creates the restlessness of many,a deep and underlying dissatisfaction, often cropping out
at the surface, with every form of inferior good, and which
imakes the knowledge of God, in and from whom is the
supreme good, like cold water to the thirsty soul. This it
is the inspiration of which all feel at times lifting them
mysteriously, like a mighty ground swell, and intimating a
connection with that which is spiritual, infinite, and eternal.
There is in this a commitment of each to himself, that each
may work out in himself the great end of all.
If to this inherent tendency and native correlation to
191
LECTURES ON MORAL SCIENCE.
good we add the provision made by God for it; if we remember that he wishes our good, and how strongly he has
expressed that wish in giving us the capacity and making
such provision for its gratification; if we reflect, too, that
the very end of God in the whole will be defeated if his
creatures decline the good for which they were made, we
shall see how sacred is the duty laid upon every man to
accept of all the good that God gives, and to seek for all
that he provides. Has he indeed given us capacities for
good, great capacities? Has he provided for their gratification? Has he made our acceptance and attainment of
that good a condition of our benefiting others? Has he
ever offered himself as the correlate of those faculties, so
that we may love him, and find our good in that love, -
what obligation can be more sacred than to acknowledge
our sense of his goodness by accepting all that he gives,
just as he gives it, and rejoicing in it?
Having thus seen that it is our duty to secure our own
highest good, let us see how, in so doing, we shall promote
the good of others, as we have already seen how, in promoting the good of others, that is, in loving them, we promote our own. When the laws of God are observed there
is no clashing of interests, but the reverse. It is a great
principle, and the gain would be immense if it could be
thoroughly incorporated into the minds of the people, that
the highest good of each man or nation is more conducive
than anything else could be to the highest good of all.
This results firom the nature of the highest good, which
is such that the enjoyment of it by one heightens instead
of diminishing that of others. It is not like a feast, of
which he who eats consumes that which might have been
enjoyed by another; but rather like a musical concert,
where each new performer, with voice and instrument ac
192
INDIVIDUAL AND GENERAL GOOD.
cordant, adds something to the harmony, and to the joy
of all. This it does, first, because, as the highest good consists in loving, and as "love works no ill to his neighbor,"
but all good, and good only, it will follow that there must
be an identity of the two. The direct tendency and result
of a virtuous love is the happiness both of the person loving and of those beloved. No one can truly love without
doing good to those beloved; and no one can truly love
without being made happy in so doing.
But, second, moral good can be sought only in and
through moral goodness. But the more moral goodness
there is in any individual, the more will others feel complacency in him, the more approve him, and the more will
he be a source of light and a ground of joy. He who
seeks his own good in moral goodness, not only lays a
foundation for his own increasing good as his goodness
increases, but also for the higher good of the whole system
with which he is connected. There is no way of doing
good so effectual as to increase our own moral goodness.
We thus increase the material of happiness, and lay the
foundation of that subtle and most efficient of all influences, an unconscious influence.
A distinction has been drawn between the enjoyment
there is from a moral act, as love, and that from the suLbsequent approbation. This is here worthy of special notice.
The enjoyment of the agent who does a good moral act is
heightened by the approbation which follows it; but such
an act is not approved by the agent alone. Whenever and
wherever a rational and moral being may become cognizant
of the act, he also will approve it, and in approval there is
joy. Approbation - moral complacency - there are few
higherjoys than from these. In them is the foundation for
the highest sympathy, and esteem, and friendship. Perhaps
17
193
LECTURES ON MORAL SCIENCE.
we do not sufficiently reflect on the immense change that
is wrought in the character of our happiness, and in all our
relations, by the addition to the natural good of good moral
acts, of this of approbation. It is that by which moral
beings have an interest in each other as such, and become
a community, a society, instead of a herd. The addition
of this element is like that of light to the heavenly bodies.
We may suppose them hanging or floating in space without light. There could then be no recogniition or watchfulness of sympathy. But let light be added, let an interchange of rays be passing throughout all space, giving to
the heavens their beauty, and proclaiming momently how
well each observes the law of its movement, and we shall
have some illustration of the change wrought by the coming in of this element of approbation and moral complacency. If we suppose a new star to be lighted up in space
we see at once that it will be in sympathy with the rest,
and that the brighter it is the more light the rest must
receive, and the more will they rejoice in it. And so it
must be with moral goodness. The more there is of it,
and so of moral good, in any one, the more material and
ground must there be for the happiness of others. "They
that be wise shall shine as the brightness of the firmament; and they that turn many to righteousness, as the
stars forever and ever."
So do we, in adopting moral good as an end, harmonize
and identify what has been called self-love with benevolence; interest with duty; the highest possible regard for
our own good with the highest possible regard for the good
of others.
From the above discussion it appears that moral good is
broadly distinguished from all other, and that its peculiarities are such that it is fitted to be the highest good. It is,
194
THE TRUE END FOR MAN.
first, independent. It is wholly within the power of the
individual. The world cannot give or take it away. It
is beyond the reach of violence or fraud. In this it is
strongly contradistinguished from all other good. Second,
't is accompanied with approbation. This may be regarded as the voice of God expressing his wish that this
good should be sought, and is in itself an entirely new
element of good. And, third, the activity required for its
attainment is identical with that which is required for the
general good. In the pursuit of it the interests of the
individual and of the community become one. To such
moral good we say, in opposition to Kant, that goodness,
or virtue, or holiness, does stand in a causal relation; and
that it is the only possible cause, since it can be only in
and from that specific form of activity.
Do we say, then, to close this discussion in the terms
with which we started,-do we say that the end for man is
happiness? No. The good here, the highest good, is fiom
the normal activity of the moral powers. As such, that
activity is obedience to the law of God, however revealed.
It is all that can be commanded or directly willed, or that
can be approved and honored. It is virtue; it is holiness.
Do we, then, say that virtue or holiness is the end for man?
No; for in this holiness there is a blessedness wholly distinctive and peculiar, higher, purer, nobler than any other;
a blessedness like that of God himself, and as inseparable from the holiness as its light is from the sun. Not,
then, in happiness without holiness do we say is the true
end for man, for without that the happiness could not be;
not in holiness without happiness, for without that the
holiness could not be and be holiness, any more than the
sun could be the sun without its light. But we do say
that the true end for man is HOLY HAPPINESS, that is,
195
LECTURES ON MORAL SCIENCE.
BLESSEDNESS. Not the oxygen alone do we need, not
the hydrogen alone, but the water,- that living water of
which if a man drink he shall never thirst.
Since, then, blessedness, as now defined, is the true good
for man as man, and so for all rational and moral beings, it
will be the end of man to increase the sum of blessedness;
and virtue will consist in a supreme purpose to promote
this impartially and in the highest degree.
The true end of man is not to be found wholly in his
subserviency to others; and it is no more to be found in
himself than the end of a stone, with its faces hewn and
fitted to be joined with others in a building, is to be found
in itself. The end of such a stone would be to be fitted in
with others as a part of the building, and if the stone were
rational it would seek as its end its own place, and would
rest there. So man, being capable of comprehending and
choosing the good of the whole as an end, is capable of
choosing his individual happiness and end in harmony with
that, as a part of that, and as being possible of attainment
only in connection with that.
But it is only through love that man is so adjusted as to
fill his place in harmony with others in a perfect society;
for, in loving, that is, in choosing the good of the whole,
man chooses his own good, and in choosing his own good
as consisting in loving, he chooses the good of the whole.
The difficulty here has resulted from a false conception
of interest. If this had always been conceived of as the
highest activity of the highest powers, there would have
been no supposed opposition between interest and duty.
The difficulty in indicating the full end for man by any
single form of expression arises from its complexity. That
he was designed to promote intentionally the blessedness
of others there can be no doubt. This he is capable of
196
THREE ELEMENTS IN THE END.
doing from a rational estimate of it, and from love. As
little doubt can there be that he was designed to be himself blessed; and he is capable of promoting his own blessedness for its own sake. Further, there can be no doubt
that in doing these he was designed to glorify God. This,
also, he is capable of doing intentionally, and this he will
do just in proportion as he promotes the blessedness of
others and is himself blessed, because the blessedness of
the creature arises from that manifestation of the perfections of God which is his declarative glory. In aiming
at either of these most would agree that man would act
rightly, and that in promoting the three to the highest
possible extent his whole end would be found. Are these
then three ends? Perhaps we may say that; still the tendency and effort has always been to find a single form of
expression that should include the three. This has been
encouraged by the peculiar manner in which these are involved in each other, since no one can promote his own
blessedness without promoting that of others, or can promote either except by glorifying God.
In the Bible form of statement the three elements are
involved. The love of God and our neighbor is made
most prominent, but when it is said, "Thou shalt love thy
neighbor as thyself," there is implied the regard for self in
just the right proportion.
In the form adopted by the Westminster divines we
have also the three elements. "To glorify God, and enjoy
him," are combined as one end. This is substantially right,
since we can glorify God only by loving and obeying hinm,
and since, in thus loving and obeying, we shall do what we
may to secure the blessedness of all other creatures.
The form adopted by Edwards is, that virtue consists in
17*
197
LECTURES ON MORAL SCIENCE.
the love of being. This includes the three elements, and,
if taken literally, much more.
In the same way we include the three elements when
we say, as above,
1st. That the end of man is to increase the sum of
blessedness, and,
2d. That virtue consists in a supreme purpose to promote blessedness impartially, and in the highest possible
degree.
3d. This being the end for which God made man, he is
glorified just in proportion as man seeks that end.
Having thus placed moral good in its rightful supremacy
as the highest good, and having found in it the point of
coalescence for individual and general good, to complete
the subject we need to see its relations to natural good.
The two are not immediately and necessarily connected.
As we have seen, moral goodness is the choosing by a free
being of his true end, together with all subordinate acts
and choices involved in that. Moral good is the enjoyment inseparably connected with such choice. It holds the
same relation to the activity of the moral powers that natural good does to that of the other powers, and is in no
proper sense a reward of moral goodness. There is in it
that which is meant when it is said that virtue is its own
reward. But, properly speaking, reward is natural good
conferred by the will of another on account of moral
goodness, On the other hand, moral badness, or wickedness, is the choice by a free being of any other than his
true end, and the acts under such choice. Moral evil,
or suffering (for it is here used for that), is the suffering
inseparably connected with such choice; and punishment
is natural evil inflicted on account of such badness or
wickedness. These, that is, moral good and evil, follow
198
MORAL AND NATURAL GOOD.
moral goodness and wickedness as the shadow the substance. Between moral goodness and a certain joy and
approbation and hope the connection is as immediate and
inseparable as any under the laws of nature, and more so;
between wickedness, the lie, the fraud, and a moral deterioration, a stain, a foreboding of evil, the connection is as
immediate and close as between putting the finger in the
fire and being burned, and more so. This effect of wickedness upon his innermost being no man can escape, and
therefore no wicked man can be, in the highest sense, prosperous. But this effect is invisible, and in this life incomplete. It is possible for a man to conceal it in a measure
from himself, and wholly from others; especially if there
be in the mind, as there commonly is, such a perversion
that moral good is comparatively disregarded, and the
possession of natural good is made the standard of happi
ness.
There is here then no antinomy, to adopt the phraseology of Kant, between virtue and happiness. If that exist
anywhere, it is between moral goodness and natural good.
Here there is, if not an opposition, yet a want of harmony
that has always given to this world the aspect of a moral
enigma. External advantages, natural good, are often
possessed by the wicked and not by the good, and the distribution of them is so far promiscuous as to jar upon our
moral sentiments, and perhaps to lead us to question the
existence of any moral government. In the oldest book
extant the inquiry is made, "Wherefore do the wicked
prosper, become old, yea, mighty in power?" More than
a thousand years afterwards the complaint was, "They
overpass the deeds of the wicked; they judge not the
cause,-the cause of the fatherless; yet they prosper."
199
LECTURES ON MORAL SCIENCE.
And the same is the complaint of to-day. Says Coleridge, -
" How seldom, friend, a good great man inherits
Honor and wealth with all his worth and pains.
It seems a story from the world of spirits,
When any man obtains that which he merits,
Or any merits that which he obtains."
Such passages show what the mind naturally regards as
moral order. It is that natural good should follow in the
train of moral goodness and wait upon it everywhere as
the satellite upon its primary. But this it does not. It is
often the reverse. Often natural good becomes the
tempter of man to lure him from virtue, and often he is
compelled, if he would be virtuous, not only to renounce
natural good, but to suffer the extremest natural evils, even
the loss of life itself. Not only does moral goodness fail
to produce natural good, -it often becomes incompatible
with it.
To relieve the jar thus made upon our moral sentiments
philosophy points us to the fact that each natural as well
as moral law is independent, and that obedience to each
gives its own separate and specific good. Be benevolent,
it is said, and you shall have the rewards of benevolence;
but if you violate the laws of temperance, your benevolence
will not and ought not to prevent your paying the penalty. The view is that men get what they earn, and that
if they do not choose to pay for a good, they should not
complain if they do not get it. Says Mrs. Barbauld, in an
essay upon Inconsistency in our Expectations, "We should
consider this world as a great mart of commerce, where
fortune exposes to our view various commodities, - riches,
ease, tranquillity, fame, integrity, knowledge. Everything
is marked at a settled price. Our time, our labor, our in
200
INCONSISTENT EXPECTATIONS.
genuity, are so much ready money, which we are to lay
out to the best advantage. Examine, compare, choose,
reject; but stand to your own judgment; and do not, like
children, when you have purchased one thing, repine that
you do not possess another which you did not purchase.
Such is the force of well-regulated industry that a steady
and vigorous exertion of our faculties, directed to one
end, will generally insure success. Would you, for in stance, be rich? Do you think that single point worth
the sacrificing everything else to? You may then be rich.
Thousands have become so from the lowest beginnings, by
toil and patient diligence, and attention to the minutest
article of expense and profit. But you must give up the
pleasures of leisure, of a vacant mind, of a free, unsuspicious temper. If you preserve your integrity, it must be
a coarse-spun and vulgar honesty. Those high and lofty
notions of morals which you brought with you from the
schools must be considerably lowered, and mixed with the
baser alloy of a jealous and- worldly-minded prudence.
You must learn to do hard, if not unjust things; and for
the nice embarrassments of a delicate and ingenuous
spirit, it is necessary for you to get rid of them as fast as
possible. You must shut your heart against the muses,
and be content to feed your understanding with plain
household truths. In short, you must not attempt to
enlarge your ideas, or polish your taste, or refine your sen timents; but must keep on in one beaten track, without
turning aside either to the right hand or to the left.'But
I cannot submit to drudgery like this; I feel a spirit above
it.''Tis well; be above it, then; only do not repine that
you are not rich."
"The man whose tender sensibility of conscience and
strict regard to the rules of morality make him scrupulous
201
LECTURES ON MORAL SCIENCE.
and fearful of offending, is often heard to complain of the
disadvantages he lies under in every path of honor and
profit.'Could I but get over some nice points, and conform to the practice and opinion of those about me, I
might stand as fair a chance as others for dignities and
preferment.' And why can you not? What hinders you
from discarding this troublesome scrupulosity of yours
which stands so grievously in your way? If it be a small
thing to enjoy a healthful mind sound at the very core,
that does not shrink from the keenest inspection; inward
freedom from remorse and perturbation; unsullied whiteness and simplicity of manners; a genuine integrity
Pure in the last recesses of the mind; -
if you think these advantages an inadequate recompense
for what you resign, dismiss your scruples this instant, and
be a slave-merchant, a parasite, or what you please."
There is good sense in this. Perhaps it is the best view
that philosophy can take; it is substantially the view of
Combe, in his Constitution of Man. But then there is in
it no vindication of a state of things in which vice so often
and so greatly gains outward advantage, and in which virtue and piety are not merely left destitute of what they
may not choose to bargain for, -to which there would not
be so much objection,- but are compelled, if they would
remain virtue and piety, to submit to the loss of all things,
and to suffer whatever the physical nature may be capable
of suffering. Of such cases the world has been full, and
for these philosophy has no solution. They point to the
future. The constitution and course of nature, with the
moral phenomena which it envelops and enshrines, does
not furnish data for its own explanation. As the solution
is not from itself, it can neither know of it, nor have organs
202
MORAL GOODNESS AND NATURAL GOOD.
to utter it. If the course of things were to go on forever
as it does now, this world, in its relation to the moral constitution of man, would forever remain an inexplicable
enigma. So far as I can judge, neither a moral government, nor a moral governor, nor the existence of any being
worthy to be called God, could be proved. No; the solution can come only from the future. This Coleridge felt;
for, while he recognizes the incompatibility just spoken of,
and so assigns to the good great man only the natural rewards of goodness and greatness, yet the friends he gives
him are such as to show that he did not suppose the solution of the problem to be here.
"What woulds't thou have a good, great man obtain?
Wealth, title, dignity, a golden chain,
Or heaps of corses which the sword hath slain.?
Goodness and greatness are not means, but ends.
Hath he not always treasure, always friends,
The good great man? Three treasures,- love, and light,
And calm thoughts, equable as infant's breath;
And three fast friends more sure than day or night, -
Himself, his Maker, and the angel Death?"
This relation of moral goodness to natural good may
doubtless be justified in a temporary dispensation. It
brings new elements into the divine administration; it
trains virtue as it could not be otherwise. It is at the
basis of moral sublimity and heroism. The object is
the enthronement of the moral nature. Let that be fully
done, and there comes the subjection, and subordination,
and right action of all the other parts of our nature, and
consequently all possible natural good from that. Here, so
far as natural good can arise from the harmony and right
action of all the powers of the individual, do we find
the natural, and, indirectly, causal relation between moral
203
LECTURES ON MORAL SCIENCE.
goodness and natural good. With the rule of the moral
nature must come all temperance, all kindness, all harmonies of the individual system, and so all the good it can
give. Here is no antinomy between moral goodness and
natural good. Naturally there is none. The present relation and arrangement is clearly a derangement, and such
an one that the moral nature can never be satisfied till the
adverse influence of evil shall be eliminated ant separated
from the good, and till external nature shall be so re-adjusted that all her substances, agencies, laws, forces, influences, shall come into accord with the laws of a higher
sphere, and shall offer themselves always and everywhere
as the servants of goodness. This, and this only, is the
natural relation between moral goodness and natural good,
and thus do we harmonize the two.
204
LECTURE IX.
THE SPHERE OF MORAL SCIENCE. - RIGHT AND WRONG. -DEFINITION OF
TERMS. - PROVINCE OF CONSCIENCE. - HOW FAR INFALLIBLE. - TWO
SPHERES. - DIVERSITY OF MORAL JUDGMENTS.- CRISES OF LIFE. - RE LATION OF CONSCIENCE TO OTHER PRINCIPLES OF ACTION. - COMPLEX ITY OF MOTIVES.-AFFECTIONS HAVE A MORAL CHARACTER IN THEM SELVES.
HAVING now examined the moral constitution, we are
in a position to discriminate more perfectly the true sphere
of moral science.
In examining an outward act of a moral being and
seeking to determine its character, we may either go backward to its source, or forward to its consequences. In
one or the other of these we must find the sphere of the
science; for though actions are often spoken of as if they
had a moral quality in themselves, yet aside from their
origin or their consequences this is not conceivable.
If we go back to the source of the act we find that
moral constitution which we have considered. We find a
person capable of doing moral acts, and of judging of
them, and it is in some mode of his activity that we find
the moral quality. In connection with this we find the
terms virtuous, vicious, goodness, wickedness, morally
good, morally evil. In connection with these there are
invisible consequences upon the spirit itself which affect
the character, and which we think of as necessary.
If we go forward to the outward consequences of the
18 205
LECTURES ON MORAL SCIENCE.
act, we find a conformity or want of conformity to fixed
relations, together with the terms utility, injuriousness,
general consequences, and more generally, though they are,
as we have seen, applied in the other direction, the terms
right and wrong. An action is good because its source is
good. "Make the tree good, and his fruit will be good."
It is right because it is conformed to a rule or law based
on a recognition of relations, and so, adapted to attain its
end. But the terms right and wrong have often been-so
applied, now to indicate moral quality as belonging to a
person, and now to indicate a conformity or want of conformity to fixed outward relations, as to produce much
confusion.
Thus it is said in the most popular work on morals published in this country,* that "Moral philosophy takes it for
granted that there is in human action a moral quality;
that is, that a human action may be right or wrong."
Here, for an action to be right or wrong, and to have a
moral quality, is the same thing. Again, in another part
of the work: "From these facts we are easily led to the
distinction between right and wrong, and innocence and
guilt. Right and wrong depend upon the relations under
which beings are created, and hence the obligations resulting from these relations are, in their nature, fixed and
unchangeable. Guilt and innocence depend upon a knowledge of these relations.".. "An action may be wrong;
but if the actor have no means of knowing it to be wrong
he is held morally guiltless in the doing of it. Or, again, a
man may have a consciousness of obligation, and a sincere
desire to act in conformity to it, and may, from ignorance
of the way in which that obligation is to be discharged,
perform an act in its nature wrong, yet, if he have acted
* Wayland's Moral Science.
206
RIGIIT AND WRONG.
according to the best of his possible knowledge, he may
not only be held guiltless, but even virtuous."
Here, then, is an act that is virtuous and also wrong.
Which, now, of these words expresses the moral quality
of the act? Virtuous, certainly. All usage would show
this; and we are also told by the same author that the
moral quality of an act resides in the intention. Here we
have the words right and wrong used to indicate moral
quality, and we have also a formal statement that they
depend upon abstract relations which have no necessary
connection with moral quality; so that an action may be
right and vicious, wrong and virtuous, at the same time.
But an investigation of "intention," on which moral quality is said to depend, is one thing; and an investigation of
"the relations under which beings are created," on which
right and wrong are said to depend, is an entirely different thing.
Which, then, of these is it, or is it both, that moral philosophy investigates? It has sometimes been one, sometimes the other; but I suppose that moral philosophy
properly stops where there is no longer any moral quality;
that moral quality is found only in mind, and that the
study of relations, and so of right and wrong as depending
on them, can be useful only as furnishing guidance for the
action of principles already formed. He who studies these
relations that he may act in accordance with them, does it
because he is already virtuous.
In a philosophy making the idea of choice and that of
an end central, the term good becomes prominent, rather
than the term right. "The True," "The Beautiful," and
"The Good," says Cousin; not, as his own philosophy
would require, The Right. Both words are indispensable,
and both are liable to analogous ambiguities, so that it is
207
LECTURES ON MORAL SCIENCE.
difficult to use either in one uniform sense. Used as adjectives, and where no moral quality is implied, the general
rule is that the term good is applied to things, and expresses that quality of the thing by which it is adapted to
the use for which it was designed; while the term right is
applied to acts, and expresses such a mode of using the
thing as will accomplish the particular end designed. Thus,
we say a good pen, meaning a pen adapted to write well;
a good axe, meaning an axe adapted to cut. But of any
use of the pen in writing, or of the axe in cutting, by
which they fail to accomplish their end, we say that it is
not right. Of any action not having moral quality, and
adapted to accomplish its end, we say, that was right.
This is the general rule, and those exceptions in which
the word right is applied to things prove the rule. Thus
a right line is that which is the most direct between
two points; the right road is that which will take a man
to the proposed end of his journey, though it may be as
far as possible from being a good one. The right man
in the right place is the man that will do the work of
that place.
When moral quality is involved, and these terms are
used as adjectives, good is applied to both persons and
actions; right to actions only. We say, a good man, and a
good act. But when we say a right act, having sole reference to moral quality, we mean the same as when we say a
good act. More generally, however, even here, the word
right, instead of looking backward to the source of the
act, looks forward to its outward consequences, and often
it is doubtful which way it was intended to look. Here is
the ambiguity.
Used as nouns, good expresses some form of enjoyment;
right is defined to mean "conformity to the perfect stand
208
CONSCIENCE.
ard, rectitude, straightness;" that is, conduct adapted to
attain the true end. With the article, as, "the Right," the
term is hardly naturalized with us yet. Generally, when
used as a noun in morals, the word right is employed in a
connection wholly different, as when we speak of rights,
and say that a man has a right to his estate, or speak of
the doctrine of rights.
Having considered the source of moral actions, and
thus the province of moral science, we next inquire after
that of conscience.
And, first, the primary activity of conscience is not
directly from the will, or what the will makes it to be. It
is not, therefore, of the nature of virtue or vice, but from
the constitution, as made by God. The fact, therefore,
that man possesses a conscience has nothing to do with
his character as good or bad. That he should have a conscience as a part of his moral nature is simply a condition
of moral character of any kind. Plain as this may appear,
the possession of conscience has often been supposed to be
a proof of moral goodness.
We inquire, secondly, how far conscience is infallible,
and so a reliable guide.
And here I observe that conscience is infallible so far as
it is uniform in its decisions. This follows from its being a
part of the constitution, or a separate faculty. That would
not be the same faculty in all men, which, under similar circumstances, should give different results. Place two men
with perfect eyes under similar circumstances and they will
see alike, and see accurately. There must, therefore, be circumstances in which there will be a uniform and infallible
action of the conscience. What are these? This will be
when the conscience is unperverted, and the subject on
which it judges is seen just as it is.
18*
209
I
LECTURES ON MORAL SCIENCE.
Under such circumstances all men would, all men now
do, immediately discriminate between benevolence and
malignity, as in themselves morally good and morally evil.
Inseparable from this idea and feeling, all would have the
blended idea and feeling of obligation, -that is, that
benevolence ought to be manifested and malignity restrained. It is by the term ought that this idea and feeling of obligation is expressed, and when our own conduct
is in question, there is in it an impulse towards the doing
of that which we feel that we ought to do. This is sometimes called the impulsive power of conscience, and it differs from others as having authority. This is the characteristic of conscience so much insisted on by Butler. It is
the proclamation within us of the moral law, carrying with
it its own authority, which no man can deny without
denying his nature. Let, now, this authority be obeyed in
carrying out the principle of benevolence, and all men
would feel approbation; let it be disobeyed, and all would
feel disapprobation. In our own case these would become
self-approbation or remorse. They would be the sense of
merit or demerit heretofore spoken of, involving an indefinite promise and threat under the divine government.
So far, and under the above circumstances, the action of
conscience would be uniform. To deny this would be to
deny that man has a moral nature, and the possibility of
moral science.
But if there be under any circumstances this uniformity, how do we account for the diversity of moral judgments there is among men, - a diversity so great that
eminent moralists, as Paley, have even denied the existence of conscience as an original part of the constitution?
To do this we must look first at the different spheres in
210
CONSCIENCE.
which conscience acts; and, second, at its liability to be
disregarded or misled.
And, first, we notice the two spheres in which conscience affirms obligation, and the different circumstances
under which it is affirmed.
Hlere it is to be observed that conscience affirms obligationII solely in view of the choice of ends, especially of the
supreme end, and not of means, except as they are conducive to the end. Conscience responds to the moral
law, and is satisfied when that is fulfilled; but the law
respects only the choice of ends. "Love," says the Scripture, "worketh no ill to his neighbor, therefore love is the
fulfilling of the law." In the exercise of supreme love to
God and impartial love to man, the law is fulfilled, and
the conscience satisfied. The means of expressing that
love must be left to positive command, or to the judgment.
What is done, then, in connection with the choice of a
supreme end is wholly in the spiritual sphere. It is in the
immediate presence of moral law. There can be no action
of a fallible understanding in estimating probabilities, and
the affirmation of obligation is immediate and uniform.
As between the good and the evil seen in themselves, it is
impossible that the moral reason should not make the distinction, and that conscience should not affirm obligation
to choose the good. To suppose otherwise would be to
deny reason to be reason. It would be to deny the possibility of conscience.
But conscience not only affirms obligation as pertaining
to the choice of a supreme end as good, but also to the
performance of acts as right, that is, as conducing to the
supreme end.
Hiere there is liability to mistake. We may, first, sup
211
LECTURES ON MIORAL SCIENCE.
pose that to be conducive to the end which is not, but the
reverse, and so approve of it as right. In this case the act
would be said to be subjectively right, but objectively
wrong. Such an act is one respecting which conscience
affirms obligation under an unavoidable mistake of the
judgment. A man may do it and be innocent. IHe may
also do such an act and be blameworthy, because he had
previously failed to inform himself as he ought. But, second, an act may, on the other hand, be subjectively wrong
and objectively right. A man may give poison with intent
to kill, that may cure an inveterate disease. According to
this, and as the word right is here used, a man may, first,
intend to do right, and do it; or, second, he may intend
to do right, and do wrong; or, third, he may intend to do
wrong, and do it; or, fourth, he may intend to do wrong,
and do right.
In the first of the above spheres, that of ends, so far at
least as the supreme end is involved, the decisions of conscience are uniform. In the second, that of means, there
is great diversity. Is this in the decisions of the conscience or of the judgment? If we suppose the decision
honestly come to that a given means is indispensable to
the attainment of the supreme end, the affirmation of obligation to choose the means will be as uniform as in the
former case. The judgment may be at fault, but there
will be no guilt. In such a case a diversity of judgment
would seem to involve a diversity in the decisions of conscience, but it would not. One man would say the thing
ought to be done, and would verily think so; another, that
it ought not; but if the decisions of the judgment were
alike, those of the conscience would be also.
It remains, then, to find the source of the diversity, and
the guilt, in some dishonesty ill forming the judgment
212
FALSE STANDARDS IN MORALS.
in some failure to come perfectly to the light, that light
which is presupposed in our being moral beings, and even
in the remorse of the wicked.
How such dishonesty may and must mix itself with
every activity of the man when once he has chosen a
wrong supreme end, we saw in the sixth lecture. That
this should ever be done is the mystery of sin. But
being done, the end becomes, of necessity, the standard
of action. To that, as supreme, everything must give
place. Now there begins a moral twilight tending to thick
darkness. In proportion as the conscience shall act, the
man must be at war with himself, and henceforth, if he
would have peace, conscience must be either evaded or
quieted. Hence the infinite subtleties of self-deception;
hence the agitations and conflicts when the conscience
will speak. Under such circumstances it is not difficult
to account for any perversion or delusion. The man is in
a position wholly false either for action or comprehension.
Hle acts in twilight, and studies astronomy from the planet
Neptune.
Thus situated, men fail to form habits of moral reflection, neither considering nor regarding what is right. They
are governed by sense, by desire, by passion, and conscience:
is held in abeyance. It is ignored. The tribunal is there,
but the cases are not brought before it.
Individuals and communities have also the power to set
up false standards which are in morals what the shrines of
idolatry are in religion. By these everything is tested, and
what they call conscience, and its vocabulary, are prostituted to the service of evil. The law is read falsely, and
-the sentence is according to the law as read. Let a man
believe that the law of God forbids his eating meat during
Lent, and if he eat it his conscience will reproach him as
213
LECTURES ON MORAL SCIENCE.
if he had committed a real crime; while the same man
may, as the inquisitors did, torture and kill heretics, not
only without remorse, but with self-complacency. There
is no fanaticism, or bigotry, or folly, that does not become
more cruel, or intense, or absurd, under the guidance of a
perverted conscience. It is this element that has given
their peculiar ferocity to religious wars, and that is apt to
make religious disputes so acrid and virulent. Certainly
men fail to come to the light, and they "put evil for good,
and good for evil."
It is plain, from the above, that to do right will be a
very different thing, as we mean by it the choice of a right
supreme end, and the determination through that of all
subordinate choices; or as we have reference to some subordinate standard, as of fashion or popularity, which we
may have adopted. According to one meaning, to do
right would be to fulfil all righteousness; according to the
other, most men can say, as they do, that they "mean to
do what is about right." This they may do in particular
instances, and often, and yet their radical character be
wholly wrong.
Of the diversity of moral judgments, then, great as it
seems, we may say, first, that there are, according to what
has been stated above, many supposed cases of such diversity that are not really such. From the complexity of the
cases presented, and the limitation of the human faculties,
men apprehend imperfectly, and so differently, facts and
their relations. Of this difference in intellectual judgment
a differing moral judgment is the result, but this implies
no want of uniformity in the action of the moral nature.
Nor, second, do the apparent vagaries of conscience when
men are dishonest with themselves, imply any want of uniformity in its action. If men will put on spectacles with
214
DIVERSITY OF JUDGMENT.
differently colored glasses, it is not the fault of their eyes
if the confusion and disputes are endless. Let them take
off their glasses and they will see alike. But, third, the
law of use and of improvement by exercise applies to the
moral powers, and under this there may be a real diversity
to a certain extent. A conscience well trained will utter
itself with greater promptness, and energy, and precision,
than one that is not.
This diversity of judgment under one guiding principle
we find in taste as well as in morals. Men are born with
some natural power of apprehending beauty. Of this they
judge; this they wish to produce. But when the question
comes to be what is beautiful in any particular case, there
are great differences of opinion. At this point it is that
the practical questions arise. Will you build a square
house or a gothic cottage? Will you paint it white or
brown? Will you lay out your grounds regularly, or irregularly, or with a regular irregularity? To such questions no original faculty necessitating the idea and the
emotion of beauty can furnish an answer. Taste must be
cultivated in accordance with principles and standards; and
such cultivation will make all the difference in decoration
and in art between the tawdriness and finery of the savage
and the perfection of taste. In connection with the study
of these principles and the application of these standards,
there will arise, as in morals, different schools of art, each
having its own merits, and gaining a supremacy more or
less wide and permanent.
We have thus, both in aesthetics and in morals, original
capacities which act uniformly to a certain extent. In
both there is an original intuition, but this was never intended to supersede the necessity for careful training.
Especially was it intended that there should be in the
215
LECTURES ON MORAL SCIENCE.
highest department of conduct, that of morality and of
duty, the highest possible combination of the intuitional
element with the results of induction, thus giving the
broadest practical wisdom. The intuitive element must
be obeyed, but what does it say? In the one case it says,
produce the highest beauty; in the other, do all the good
you can. For these no consultation is needed with any
one, and no advice. But when the practical question
comes whether a young man engaged in mercantile business shall give it up to prepare for the gospel ministry,
something more is needed. Here the inductive element
comes in. It will depend upon his age, his talents, his
means, upon those dependent upon him, or likely to be;
and upon this he may properly ask advice. Of this kind
are most practical questions. Nothing can be more trying than the suspense and nice balancings these often require; and when the decisions are made, they will be
those of beings limited and imperfect, liable to mistake
even where there is no sin. Respecting decisions of this
kind men need much mutual forbearance in judging of
each other. It is seldom that we can put ourselves fully
in the place of another, and no general rule can be laid
down. We can only say, with the wise man, that "wisdom is profitable to direct."
In connection with these two spheres of judgment, the
intuitional and the inductive, I would call your attention
to the two great crises in the life of every young man,of most persons, indeed, who come to maturity,- and to
the very different character of the elements and questions
they involve.
The first and great crisis is that which involves the
whole of duty and of destiny under the government of
God. In the life of every one much instructed a point is
216
CRISES OF LIFE.
reached when the question consciously arises respecting a
supreme end. It is found that a wrong one has been
chosen, -shall there be a change? The question here is
between two things different in kind and utterly incompatible. It is one of giving and receiving. Shall the person give himself up in love to the service of God and
man? or shall he regard God and men as means through
which he may receive what he desires? The question
may not be thus stated, but it involves this; and there are
no balancings, and agitations, and suspense, like those often
connected with its decision. Here the intuitional element
and the will are alone concerned. The question is not one
of means, but of ends. It is aloof from the relations of
time. There is no place for induction or call for advice.
Neither is there room for doubt. Intuitively, peremptorily, persistently, the conscience affirms obligation. It is
now proximate to the will, and these are like two vessels
grappled in conflict. They are the Monitor and the Merrimac. The question is one of simple obedience. Will
the will yield, or will it not? Will the man come into
harmony with himself and with God, or will he not?
This question no one can decide for another, and the act
required is so simple and elementary that no one can tell
another how to do it. To attempt this, as is often done, is
like attempting to define an elementary notion. This, if
the question is to be decided once for all, is the crisis not
only for this life, but for the whole of existence.
The second great crisis in the life of a young man comes
when he is to decide on his profession, that is, on the particular form and direction of his activity under the general
choice previously made. On this will depend not merely
the amount, but the kind of good he will do, the books
19
217
LECTURES ON MORAL SCIENCE.
he will read, his professional friendships, the line of his
thoughts, and the principles of their association.
The question here, it is often supposed, is to be determined by conscience; but if the previous question has
been fully settled, conscience has, in strictness, nothing to
do with it. The simple question will be, in one case, how
we can do the greatest amount of good; and in the other,
how we can best subserve our own private ends. These
are questions of comparison and judgment involving many
particulars. Men equally conscientious might decide them
differently, and with them conscience has nothing to do,
unless it be to secure for them a careful and candid attention. The process is like that by which we find the minor
proposition of a syllogism. We inquire whether a particular proposition comes under another that is more general.
The distinctions above made will enable us to account
in part for the confusion there has been in our moral philosophy, and particularly for the prominence given to right,
and the right, as distinguished from the good.
The choice of a supreme end is generic. It is made
once, in a sense only once. In a sense, too, it is made
always, constantly repeated, since it is only under this that
other choices are made. It is like the light of consciousness, and would naturally be the last thing investigated.
Indeed, as consciousness is the generic form of intelligence, and the desire of happiness that of the desires, and
love that of the affections, so the choice of a supreme end
is the generic form of volition. It enters into all the
others; they are made in its light and partake of its character. In respect to this the affirmation of obligation is
not constantly repeated in any specific form, and may be
scarcely thought of for years. But the affirmation of obligation as connected with right, or what is supposed to be
218
CONSCIENCE AS A MOTIVE.
so, is constantly made, as it is concerning that that practical questions and discussions constantly arise. Hence it
has attracted the chief attention. The immediate question, and that on which the obligation would turn, has
been one of right, and hence the idea of right in its relation to obligation has been supposed to be ultimate. It is
ultimate only as the seaport is ultimate where everything
is stopped, examined, exchanged, but which would be no
seaport at all but for the ocean beyond. If there were
nothing good, no end to be chosen, there would be nothing right. The only question is whether the right is the
good.
WVe next inquire after the relation of conscience to the
other active principles. How far should it be merely regnlative, and how far a positive principle of action?
A philosophy of ends requires for the person that which
we find in all nature below it, a good which shall result
from the congruity of itself with that in which its good is
found, and which shall come immediately from the activity
of the one in its relation to the other. Everywhere there
is duality. In vegetable life there is the living seed, there
is moisture, air, and warmth, and there is growth. In sensitive life there is the eye, and there is light, and from
these, vision. Not from the eye alone, or from light alone,
does seeing come, -but from the two in right relations.
Seeing is not a thing, a being, but a product and result of
vitality acting according to its laws. Of that inward constitution and congruity by which the eye and light are
adapted to each other we know nothing. We only know
the facts, and the conditions, or laws of the facts. So
again in the appetites. There is hunger, and there is
food. The enjoyment is not from the appetite alone, or
from the food alone, but from both in the right relation.
219
LECTURES ON MORAL SCIENCE.
In the mental world there is the knowing mind and the
object known. Knowledge is not from the mind without
the object, nor from the object without the mind, but is
the result of the two in the relations intended by God.
The same holds in the affections and the will. There
is the love and the object loved, the will and the thing
willed, the choice and the thing chosen, and the blessedness is not the result of the mind alone or of the object
alone, but of the right relations of the two.
And here, as we reach enjoyment from love, I wish to
notice a peculiarity, which, according to our previous principle of classification, must place that higher than any
other, at least when it is in its fulness. It is higher, not
merely as the product of our highest powers, but as more
complex. It has the element of reciprocity. It is not
simply because personal beings are higher and intrinsically
more excellent than others, or that we can have affections
for them specific and peculiar, that the activity of our
faculties when they are the object can give us a higher
joy, but because they are capable of the conscious recognition and reciprocation of that affection. Hardly less than
the joy of loving is that of being beloved. Here we find
the necessity of each for all, and of all for each, and the
foundation for the highest good of all and of each. The
highest conceivable good must be from a conscious and
perfect accordance of the will and moral affections with
those of a being of infinite excellence who should recognize and reciprocate the affection.
In all this it will be observed that the relation of the
faculties and their objects is immediate and direct.
But with such a constitution in perfect adjustment it is
plain that the office of conscience, if required at all, would
be simply regulative. We have provision here for activity
220
OFFICE OF CONSCIENCE.
throughout its whole range, without the conscience. We
are not to eat from conscience, else why the appetite?
The affections are not to act from conscience, else they
would not be original parts of our nature. It is not the
office of conscience to supersede any of the natural prinples of action, nor can it ever lead to action except as
there are grounds for that action furnished by principles
other than itself. In this respect it is analogous to selflove. Its office is to affirm obligation to clioose in a particular way. The grounds of choice are presupposed.
Hlence the double function of our moral nature. As a
condition for the action of conscience man has the power,
in virtue of his moral reason and of the affections growing out of it, to apprehend the end which he ought to
choose, and he is drawn towards the beings whom he ought
to love. This end he apprehends and chooses, these beings
he appreciates and loves, as a moral being, not merely from
a sense of obligation, but from their inherent worth or
excellency. Acting in the light of moral reason, from the
play of the moral affections, and in the exercise of freedom, man is a moral being; but as there is an alternative,
and he is liable to choose wrongly, the moral nature would
not be complete if there were not an affirmation of obligation to choose the good apprehended by the moral reason,
and towards which the affections were drawn. But conscience is that function of the moral reason by which it
affirms obligation to choose primarily tile good, and secondarily the right, from its apprehended relation to that.
This is its function before the act. Subsequently there are
evolved its rewarding and punishing, and its prophetic
power.
We have, then, as a condition for the action of conscience in affirming obligation, not the mere perception by
19*
221
LECTURES ON MORAL SCIENCE.
the intellect of external relations, but those prim,l acts of
affection and choice which are involved in our very conception of ourselves as acting, and which spring from that
which is deepest and most sacred in our nature. As
means of fulfilling the obligations thus affirmed when actions other than the mere choice are required, we have the
perception of actions as right, sometimes apparently intuitive, but often clearly the reverse, and in which we are
liable to great mistakes.
We may now see when the action of conscience is
requiied; when it is merely regulative, and when it becomes a direct impulse to action. Let a man, as has been
said, eat from appetite. This is a sufficient reason for his
eating. His appetite was given that he might eat. There
will be no need of conscience here, except as assenting,
or as there may be room for doubt respecting the quantity or quality of what is to be eaten. On both these
points a man isoound to use his best knowledge, and if
there be reason to suspect anything injurious, conscience
will say no. It will become the veto power of the mind.
Self-love may forbid it as opposed to individual interest,
but conscience will pronounce it wrong as opposed to the
great law of love.
If we might suppose the appetite so constituted as to
crave nothing injurious, there would be no need of supervision by either self-love or conscience, and such need
must be in proportion to the want of autonomy in the
appetite. The above is the law for all natural appetites
and principles of action that are sufficiently strong to
induce all the action in their line that is required by their
ends. The office of conscience will, then, be simply to say
no, if there be a tendency to excessive or perverted action.
But if any faculty or impulse, as the desire of knowledge,
222
CONSCIENCE- RESTRAINING AND IMPELLING. 223
be not strong enough, then it may and should be put at
work under the imperative of duty. Then conscience
not merely says no, - it impels; it becomes a task-master.
In such cases, however, we feel it to be unfortunate and
undesirable that the action of conscience should be needed.
The faculties do not work cheerfully, but as under a yoke.
We prefer that a sense of duty should be required to
restrain a boy from study, rather than to impel him to it.
This holds true of all the instrumental powers. They
have no moral character, except as the man accepts and
directs them, and we prefer to see them full and redundant,
rather than meagre. But when we reach the directive,
and especially the moral powers, the relation is different.
Here conscience does not come in as a power from without, standing above the faculty, but as part of a circle of
activity that is necessary for the completeness of the whole.
In all life there is a circle, every part of which implies
every other. We may treat of respiration separately, but
it involves circulation and digestion, and there is no life
without each. So in the moral life. We may treat of the
apprehension of good and the power of choice apart from
conscience, but all are necessary to life. Hience, in a natural, that is in a right state, a sense of obligation will so
coalesce with moral affection as not to produce constraint,
but rather to heighten joy. It will be as the seal of a bond
to which we set our names with an unwavering confidence
and an ineffable delight. Between the affections and the
conscience there will be full consent, and there is not a note
in the harmony that goes up from the full action of the
moral powers that is more pervading and would be more
missed than that from an approving conscience. It is in a
spring of affection, coordinate with the affirmation of obli
LECTURE$ ON MORAL SCIENCE.
gation, that we have a marriage of strength and beauty,
whose fruit is blessedness.
That the moral nature does not, like the other powers,
work under a yoke in the presence of obligation, but
simply finds its own completeness, it is important to see,
because there is a prevalent impression that there is in all
obligation something of constraint. It is an objection to
the system that makes the right ultimate, that, as based on
a mere abstraction, it furnishes no object for the affections,
and moves us through its imperative by constraining and
driving, rather than by attracting us. In our conception
of a perfect being the law is not known as an outward
and constraining force, but there is a coincidence with it
of inclination and of will by which perfect obedience
becomes perfect freedom. Love is free and directly from
a view of its object; but "love is the fulfilling of the
law."
Seeing thus the relation of conscience as a motive to
the other active powers, let us look at the gradation and
possible complexity of motives in human action. The
gradation of motives will follow from that of the powers,
and will be in accordance with it. Motives will be higher
or lower as the powers are from which they spring, and
both virtues and vices will be designated from the sphere
of activity that is rightly directed, or abused. A man
abusing the sensitive part of his nature is sensual; rightly
using it, he is temperate. Abusing the desire for power,
he is selfishly ambitious; rightly using it, he is beneficent.
Thlroughout, both virtues and vices are designated by the
activities employed. If there are not high vices, there are
those that are low, and some virtues are higher than
others.
Of the possible complexity of motives we may gain a
204
COMPLEXITY OF MOTIVES.
correct view if we suppose a father to command his child
to get a lesson. IHere there is, first, between the mind of
the child and knowledge a natural congruity, and he may
get the lesson from a simple love of the knowledge. It
may be just what he would have done without any command. But the parent says further,- If you get the lesson it will aid you in getting a living; you shall have my
approbation, and I will express that by a reward. If you
do not get it you shall have my disapprobation, and I will
express that by punishment.
Now, between the mind of the child and the approbation of the parent there is a congruity. It is right for him
to desire that approbation. The reward itself may be one
that would appeal to the legitimate desires and which it
would be right to seek, and the reverse may be said of the
disapprobation and the punishment.
We have, then, as motives, 1st. Love of knowledge for
its own sake. 2d. A desire for it as useful in gaining a
living. 3d. A desire of the reward. 4th. Fear of punishment. 5th. Regard for the authority of the parent. 6th.
A love of his approbation. 7th. Dread of his disapprobation. 8th. The affirmation by conscience of obligation.
Of these each is legitimate, -is appropriate to a rational
being, - is right. Each may take its turn, or they may
conspire together; and if from any one or all of them the
lesson should be learned the parent might be satisfied.
Still, it must be remembered that the ultimate character of
the mind in every movement relative to these motives will
be determined by that generic act of choice under which
they all take place. The motives may be objectively right,
but the man not subjectively right in being governed by
them.
If the preceding remarks, or indeed the general doctrine
225
LECTURES ON MORAL SCIENCE.
of these lectures, be correct, they will go far to determine
the question whether in order to be virtuous an act must
be done from a sense of duty. On this point distinguished
thinkers differ. Chalmers says this is essential, and, as is
usual with him, reiterates and enforces the point in a variety of ways. "It is not," says he, "volition alone which
makes a thing virtuous, but volition under a sense of duty;
and that only is a moral performance to which a man is
urged by a sense or feeling of moral obligation." Again
he says, "Whatever cometh not of a sense of duty hath
no moral character of itself, and no moral approbation due
to it." This opinion of Chalmers is quoted with approbation by MeCosh. On the other hand, Dr. Woods says, "It
would be very easy to show that moral affection may exist
in one who has at the time no distinct apprehension of its
nature, and no present feeling of approbation or disapprobation." "I say, then," he continues, "it is not essential
to our moral agency, or to the existence of moral good
and evil in us, that we should at the time have a distinct
consideration or conception of a moral law, or a sensible
approbation or disapprobation of our feelings and actions."
This inquiry runs back to the constitution of the moral
nature as involving any other element than that of right
and of conscience, and to the question whether there is
anything virtuous in the moral affections and the will
when they act according to their own law, and directly
with reference to their objects.
That there is something thus virtuous, McCosh, in opposition to his direct assertion, seems everywhere to imply.
Thus he says, "Much of human wickedness is displayed in
the ingenious schemes which are contrived to deceive the
moral faculty and avoid its humbling judgments." This
implies something having wickedness, and yet acting inde
226
INHERENT VIRTUE.
pendently of the moral faculty. Again, he says, "Moral
excellence is truly the whole powers and affections of the
soul in healthy exercise, and in order to guard it there is a
faculty with a train of corresponding feelings, presiding
over all the other faculties and seated in the very heart."
This implies excellence before it can be guarded. Again,
he says expressly that "the moral quality is not given to
the action by the mind contemplating it." "It is not our
perception and approbation that renders a benevolent action good, but we perceive its excellence and approve it
because it is good."
It is not to be supposed that moral actions are done
except under moral law and some generic choice of good
or evil. But as a vicious man does not do evil actions because of their viciousness, so neither would it seem necessary that a good man should do good actions because of
their virtuousness; at least it cannot be implied, as it
seems to be in the statement of Chalmers, that a sense of
obligation is the only virtuous motive. It is the law of
the affections that they are drawn out in view of their
object. An interested love is impossible. Only from an
apprehension of some quality in the being loved can love
come. The love of God may imply virtue. It does so.
But the love cannot be in view of its own virtuousness, or
with the thought of that, but must be in view of either
the worth or the worthiness of God.
227
2.#
LECTURE X.
RECTITUDE AND VIRTUE. - RELATIONS.- EXPEDIENCY, PRUDENCE, AND
VIRTUE.-ORIGIN OF MORAL DISTINCTIONS AS RELATED TO THE DIVINE
NATURE. - COINCIDENCE OF INSTINCT AND REASON - OF FAITH AND
REASON - OF PHILOSOPHY AND RELIGION.
IT will be remembered that in observing a moral act
we went backward to its source. In so doing we found a
moral constitution. That constitution we have examined,
and have found its end and law. In doing this we considered those voluntary states of mind which are in themselves good or evil, virtuous or vicious.
We know immediately and intuitively that love is good,
and malignity evil; and it is inconceivable that their nature should be changed by any will. They are opposites,
as are light and darkness, hardness and softness; one may
give place to the other, but can never become the other.
This, I suppose, is what is meant by the eternal and immutable distinctions of morals.
It was then said that in passing outwards from a moral
action we found right and wrong, utility, expediency, general consequences. It will be next in order to examine
these, and to inquire how far their claims may be reconciled with those of virtue without confounding the two.
As has been said, right is often used as synonymous with
virtuous, and wrong with vicious. The right, also, seems
to be used as synonymous with moral goodness; at least
if that be not its meaning I am unable to say what it is.
But by right is also meant conformity to a rule or law,
228
NATURAL GOVERNMENT.
tendency to an end, accordance with fixed relations, and
by wrong, the reverse. "Right and wrong," says Dr.
Wayland, "depend upon the relations under which beings
are created, and are invariable." In this sense actions
may be right or wrong without reference to the character
or intention of the agent. In the first sense they cannot,
and the trouble has been that these terms have been used,
now to indicate virtue as originating in will, and now to
indicate a quality, sometimes called moral, that has no reference to intention. So far as they are used in the first
sense we have already considered them, or rather that
which they indicate; it is in the second sense that they
now claim our attention.
Plainly the results of human conduct in this life are not
determined solely by the dispositions and intentions from
which they spring. We live under a natural as well as
under a moral government, and the first is the instrument,
frame-work, and prophecy of the second. We are surrounded by other beings, and by an external nature that is
complicated, involving numerous substances, and forces,
and laws. These beings, this nature, these substances, and
forces, and laws, have a determinate constitution in accordance with which we may act upon them and they upon
us,.ii( this action, at least so far as nature is concerned,
will not )c affected by our state of mind as good or evil,
orby any intention that may spring from that. Between
us and external nature there are fixed relations, and the
result will depend upon our acting or not acting in accordance with those relations.
A being wholly virtuous may act in entire accordance
with the nature of the beings and substances around him,
and then the whole result will be right and good. Again,
with character unchanged, but ignorant of the relations in
20
229
LECTURES ON MORAL SCIENCE.
which he is placed, the same being may so act as to produce suffering to himself, or others, or both. He may
intend to preserve his health, but be ignorant, and unavoidably so, of the effect of a want of ventilation, and in
consequence may live for years in debility and suffering.
Such a person would not act in accordance with "the
nature of things;" or, as it has sometimes been expressed,
with "the fitness of things;" or, as it has been expressed
again, with "the truth of things;" or, once again, with
"the relations in which he was placed."
From the very nature of man it is impossible he should
act except in some relation. Hence the consideration of
relations -not merely of things as they are in themselves,
but in their relations- must always enter into our estimate both of propriety and of duty.
So numerous, indeed, and complex are these relations,
and so intimately is their right adjustment connected with
human well-being, that not a few moralists have supposed
moral obligation, and so the whole science of morals, to
be founded on them. "It is fit," says Dr. Samuel Clarke,
"that man should obey God, and therefore he ought to
obey him." It is true, according to Wollaston, that fresh
air is needed for health, and he who acts as if it were not,
acts a lie, and therefore does wrong. "The relation of parent and child," says Dr. Wayland, "is constituted by God,
therefore men are bound to act in accordance with that
relation." He asserts, moreover, that the sense of obligation arises immediately on the perception of the relation.
But it does not seem to be true, it may be observed
here, that a child is bound to obey his parent simply because he is his parent. A parent may be an idiot, or
insane, or intoxicated, or wholly abandoned to vice, and
then the law makes provision for the guardianship of the
230
RELATIONS.
child, that is, for placing him where he need not obey the
parent. The sole reason why the child is to obey the
parent is the presumption that the end for which God
made him will be thus best secured. If it could be certainly known that that end would thus be defeated, the
child would not be under obligation to obey, but the reverse. Relations cannot indicate what is good. They
may, and do, what is right, but so far as they do this they
may all be reduced to one, that is, the relation of any act
which a moral being may be required to do, to his end.
So absolutely is the will of God revealed in that, that it is
inconceivable he should lay a being under obligation to do
anything not in accordance with his highest end.
But while we do not find the foundation of morals in
the nature, or fitness, or truth of things, or in any mere
relations, we may not overlook the important part which
a perception of these was intended to play in the regulation of human conduct. Not only, as has been said, may
a virtuous man fail to conform to the nature of things, or
to the relations in which he is placed, and thus suffer; but
a man not virtuous may conform to them, and be rewarded.
Beneficial effects will follow without respect to the motive.
There is a sense in which an action thus conformed to the
nature of things is right although the motive may not be
good. It is, as we say, right in itself; it is conformed to
the nature of things; it is fit, and suitable, and proper, and
what ought to be done. Let a man be outwardly honest;
let him pay his debts, and tell the truth, and though he
may do it simply because he thinks honesty the best policy,
and so not be virtuous, yet the acts are right in themselves,
and the confidence of men in each other and the prosperity of the community will be promoted by them. On the
other hand, from an imperfect apprehension of relations, a
231
LECTURES ON MORAL SCIENCE.
person may take the redress of his private wrongs into his
own hands, or may buy and sell lottery tickets, or intoxicating drinks, or perhaps be a polygamist, and while the
motive may be good, that will not prevent the disastrous
effects of these acts as wrong in themselves. Sooner or
later such acts will work out their own retribution.
It cannot, indeed, be too clearly seen that into the whole
system of nature as related to us, into the human constitution in its very texture, into the constitution of society,
there are not only inwrought laws of reward for conformity to relations and fixed laws, but also laws of retribution
that seem to execute themselves. Violate a law of nature
by stepping from a precipice, and you fall; violate a law
of your organization by intemperance, and your punishment will be in proportion to the offence. And so of
society. Violate the law of its organization as one whole
so that portions of that whole are neglected and degraded,
and that very violation will work out a sure retribution.
In all this we see only the working of fixed law without
regard to motives or character. A mistake is punished
just as severely as a wilful violation of the law.
There is that in the working of these laws that is precisely as if there were a moral instinct in all these departments. As with instinct, let everything else be as it
should, and these laws will work right, and produce only
good; but let there be perversion and derangement, and
then, like instinct again, they will work blindly and disastrously. They may not overturn, but will utterly disregard all moral distinctions.
If, now, we carry the working of these laws into the
mind, we shall have the whole of what many believe to be
the moral system of the universe. They believe there is
no reward or punishment except from the operation of
232
RESULTS OF GUILT INCALCULABLE.
fixed laws, and know nothing of a personal being working
within or through those laws. Especially do they know
nothing of one working above them.
The difference between those who do, and those who do
not recognize a personal being at this point is radical, and
forms a dividing line, not only between schools, but between individuals in their habits of thought who know
nothing of schools. To me the indications of a moral system in these laws are like the indications of reason in
instinct, or rather lying back of it, and needed to account
for it. Taken by themselves, they are a mute, imperfect,
often baffled expression, not so much of the thing itself
as of a yearning after it. They are the harbinger of the
thing, a beautiful frame-work into which a perfect virtue
may be fitted. But a system thus regardless of character,
and as cold and remorseless as that of fate, can never meet
the demands of either the conscience or the affections.
Of acts under such a system, whether right or wrong, the
good and evil results may be calculated, because they are
wrought out within and by the system, and will be the
same, whether intended or not. But the results of virtue
and vice cannot be calculated, because they depend immediately upon will, and involve the principles of a moral
government that has an extent and bearings wholly beyond our comprehension. Here, as well as in the absence
of approbation and disapprobation, is a great difference
between this system and one truly moral. When a rational being wilfully goes against the laws of that being;
when the child refuses to obey his father; when the creature knzowingly disregards the will of the Creator, it is
impossible to say what may be the results. Hlere we find,
not merely a mistake, which is an intellectual crime,not merely ar want of acting in conformity with fixed and
20*
233
LECTURES ON MORAL SCIENCE.
impersonal law, but guilt, -the most dreadful word and
thing, and the only thing to be really dreaded, in the universe of God.
But while we cannot identify what is right with what
is good, they are yet closely allied. Goodness is good in
itself. Regarded as a fundamental choice conformed to
moral law, it is also right, and must involve a disposition
that would lead to the doing of all right acts. As tending, then, to produce right acts, goodness is righteousness; and if there be adequate knowledge, there will be
in all its acts rightness,-that is, an entire conformity to
all fixed relations, and so to all divine law. Nothing but
sufficient knowledge can be wanting to effect a perfect
coincidence of all virtuous and of all right action. An
action will not be right because it is virtuous, nor virtuous
because it is right. For the one we look backwards to its
source; for the other, forwards to its relations and consequences; but virtue cannot be virtue except as there is in
it a disposition to do right.
If, then, any contend for an absolute and ultimate right
that is identical with goodness, we are content; but if not,
it will devolve on them to show what they mean by it that
is different from the above statement.
Having thus treated of right in its relation to virtue, we
next inquire after the relation to it of utility and expediency. This has been a difficult point in morals; but those
who accept the above statements will readily see what that
relation must be. Whatever is useful or expedient must
be so with reference to some end. Hence, utility and
expediency always imply an end previously chosen. Here
nothing will be chosen for its own sake, and all questions
must respect the choice, not of ends, but of conditions and
means. We have thus two classes of questions closely
234
MEANS AND ENDS.
connected, but of an entirely different order. An ultimate
end, so far as it is regarded as ultimate, - and a supreme
end must be wholly so, - can have neither utility nor
expediency. It is chosen for its own sake. In the choice
of such an end character alone is involved. In that of
means and conditions, character is always implied; but it
is capacity that is chiefly involved.
And here it is to be observed that conditions and means
can be rationally chosen only on the ground that they are
adapted to the attainment of the end; and that, having
chosen an end, it would be an inconsistency and folly
not to choose those conditions and means that would be
the best adapted to attain it. What else can a rational
man do?
It is to be observed, again, that if the end chosen be the
true supreme end of man, then any means in themselves
adapted to attain that end will be right. This is not the
doctrine that the end sanctifies the means, but implies the
fact that this is such an end as can be obtained only by
sanctified means. An inadequate and false end, chosen
selfishly, may be attained by vicious means. If money be
the supreme end, it may be attained by fraud. Power
miuay be attained by violence and injustice. When a man
has chosen a supreme end, if he be consistent, he will use
whatever means may be necessary to attain it. The very
fact that the end is supreme to him will render it impossible that anything should come between him and it.
But no man can seek to promote blessedness by sin,- by
any interference with the rights of others. It would be
a contradiction.
In treating of the intellect, it was noticed how the
wrong choice of a supreme end will pervert that. We
now see how it is that it will pervert the heart; and how
235
LECTURES ON MORAL SCIENCE.
it is that no man can consistently stop at any wickedness
necessary to the attainment of such an end. If the love
of power be absolutely supreme, then the esteem of men
and the favor of God must be relatively of less account,
and must be disregarded. And as we saw how the choice
of the true end would lead to candor and a coming to the
light, so we may now see how it involves unselfishness in
the choice of means. We here find deep and pervading
laws of our constitution.
We may now see what the true doctrine of expediency
and utility is, and how largely the consideration of these
must enter into human life. In all secondary choices and
executive volition they must govern every rational man.
Does a man pursue his own ends by what he deems to be
the best means? If not, he is blameworthy. But if he
does, it ill becomes him to rail at expediency.
But may not expediency be opposed to right? A false
expediency may; and hence the prejudice. The objections to expediency seem to have arisen, first, from the
general choice of wrong ends, and measuring expediency
by them. Two men have a quarrel. The object of each
is to humble the other. One can do it by fraud. He says
this would be expedient, but not right. Expedient for
what? Not for the promotion of blessedness - on the
whole. And, second, objections have arisen because, even
when the true end has been chosen, men have sometimes
failed to see that they might not and never could promote
it by interfering with the rights of others, - that means,
to be means at all, must here partake of the character of
the end.
This true doctrine of expediency some have failed to
see; and in seeking, in opposition to that, for an abstract
right, have fallen into a fanaticism not the less mischievous
236
PRUDENCE AND VIRTUE.
for its high pretensions, and the fair garb in which it has
clothed itself.
We now pass to another distinction which belongs here,
- that between prudence and virtue. Prudence does not
furnish positive motives. It presupposes the choice and
pursuit of an end, and its office is to guard against danger
under the operation of fixed laws. "The prudent man
foreseeth the evil, and hideth himself." Prudence has sole
reference to our interest as under fixed laws, and not as
under the rule of a personal being. In its perfection it is
an acquired sagacity in regard to the laws, whether of
matter or of mind, and a habit of shaping the conduct in
accordance with them; and as such, it becomes to man in
his sphere what instinct is to the animals in theirs.
As prudence regards only consequences from fixed laws,
the moment we come to faith, and to suffering and martyrdom for adherence to principle at all hazards, its sphere is
transcended. A man may purposely so act in opposition
to fixed laws as to jeopardize or destroy his whole interest
under those laws; and this he may do rationally, but in no
proper sense of the word can he be said to do it prudently.
The hero is not prudent. The martyr is not prudent. He
is brought into a position where the rules of prudence are
out of place, and where it becomes necessary to vindicate
ihe supremacy of the spiritual nature and the majesty of
virtue by an unconditional trust in goodness and in God.
Prudence may then be regarded as the appointed guardian of the interests of man in the present life, and under
those fixed laws that are made known by experience. As
such, it is, as Butler says, "of the nature of virtue." To
study those laws, to heed them when we may, and to
secure the good there is under them, is our wisdom and
duty. Wantonly or heedlessly to disregard them is wrong;
237
LECTURES ON MORAL SCIENCE.
but the highest life of man is not in them, or from thenm,
or under them; and when the demands of that life require
us to give up every interest under those laws, or even to
lay ourselves down in their track to be crushed by them,
it is to be done.
Right, expediency, prudence, are all found by tracing
actions outwards, and have their basis in a nature of
things, and in which some have placed the foundation of
morals. Of this nature of things, there are some who so
think as if it were something back of the will of God, and
controlled it. So Dr. Dwight. HIe says, "It is, I apprehend, evident that the foundation of virtue is not in the
will of God, but in the nature of things." And to this
nature of things he supposed the will of God to be conformed. He says further, confounding, as it would seem,
the nature of things with their tendency, that "virtue is
termed good only as being the cause of happiness," so
making the foundation of obligation to be in the tendency
of things, and the will of God to be governed by that,
as if there could be either a nature or a tendency of things
that did not have its origin in the will of God.
Others, again, supposing the nature of things, and so
their tendencies to be originated by God, go back to the
nature of God himself for the origin of these distinctions
and the foundation of moral obligation. Here they seem
to find the limit of all analysis, and of all thought, since
nothing can be more ultimate than the nature of God.
"Instead of any abstract fitnesses being the standard of
the divine nature, the divine nature must itself,' it is said,
"be the origin and standard of all fitnesses."
On a subject like this it becomes us to speak cautiously
and reverently. It may be doubted, however, whether
this mode of speaking either originates in, or conveys a
2 OD 8
PURE PERSONALITY OF GOD.
true conception of the nature of God. It supposes a
nature in him that lies back of reason and of will, and,
from which impulsions come by which his will is necessarily determined. Because we have a nature that is distinct from our personality, and underlies it, it is imagined
that God has. It may be, however, that the nature
of God is nothing distinct from his personality, and that
so he is wholly supernatural. It may be that the terms
nature and natural, used as they commonly are to indicate something fixed, stated, uniform, and not made so by
will, are without meaning when applied to God. So far
as we can apply the term to God, it may be that it is his
nature to be simply a Person determining his own will
in the light of an all-comprehending reason, and in view,
not of any intrinsic differences in a nature of things, but
of the different character and results of different possible
forms of his own activity. It may be that what we must
reach in our ultimate analysis is a free personality, -a
Person, with no nature, or fate, or fitnesses of things back
of him, or above him; who is himself, by his own firee
choice, the originator of everything that may properly be
called nature, and of all fitnesses of things. That this is
not so, who shall say? Who shall say that this is not our
only way to avoid that conception of God, so very general,
that is equivalent to fate?
The confusion at this point may be largely due to the
inadequacy of language for such a subject. Two extremes
were to be avoided: one, the founding of obligation on
mere will; the other, the virtual exclusion of will.
The difficulty arises from the eternity, and so the necessity, of the divine existence. But if it be said that God is
a necessary being, it is also to be said that he is necessarily
rational and free, and that what he is now he has always
239
LECTURES ON MORAL SCIENCE.
been. Go back as we may, we find simply a personal God,
rational and free. As such, he is a law unto himself, subject to no necessity, except the necessity there is that
reason should act rationally. Obligation is affirmed in
him, as in us, we being made in his image, only with no
danger of mistake, and with no possibility that he should
be responsible to any one. This gives us, as the origin of
all things that had an origin, character and will; and
instead of a blind fate it gives us that moral certainty
which accompanies the highest freedom. Something must
be given. What we need is simply a person; and it is a
mere abuse of language to convert that constitution of the
Divine Being, by which he is a person, and capable of a
rational freedom, into a nature the very idea of which
excludes freedom.
But if this be so, then, as in our search backwards for
the origin of being, the ultimate fact is the being of God;
so, in our search backwards for the origin of moral distinctions, we shall find, not any nature of things, not any
nature of God, not any necessary and eternal principles,
but simply the character of God. It would be a grand
consummation thus to find, standing at the termination of
all our investigations physical and moral, as that beyond
which nothing could be more ultimate, simply the being of
God, and the character of God.
If this be so, then virtue or goodness, and rectitude,
will, in God, be the same thing seen in different aspects.
Ilis goodness will be seen in his choice of ends, and his
rectitude in the mode of attaining them; and there can
be for man, and indeed for any creature however exalted,
nothing higher, or better, or more ultimate than conformity
to the character of God. It may be that, as all the natural
teachings of the works of God are but indications and
240
THE WILL OF GOD A STANDARD.
expressions of his natural attributes, so all their moral
teachings, together with those of revelation, are but the
expressions of his moral character; and that the end of all
teaching, and of all influences, will be the formation by
creatures made in his image of a character similar to his.
If we accept what has now been said, it will follow,
as moral distinctions have their origin in God as a person,
as his character is the standard of goodness, and his will
is the expression of his character, that his will, however
made known, must be the ultimate rule of moral action;
it must be that to which the conscience will respond, not
simply as will, but as the will of God. It was made to
respond to his will because that is the expression of his
character; and his character, as combining benevolence and
rectitude, is the perfection and standard of moral excellence.
As we, then, find in the being of God the origin of all
other being, so that without him there could be no other;
so do we find in the character of God, and in his will as
expressing that character, all that is ultimate in moral distinctions, and without that will and character those distinctions could not be. Thus do all our speculations lead
us to God, not merely as the fountain of being, but of
excellence, and as the Head and Governor of the moral
universe.
We have now examined the human constitution as related to ends rationally apprehended and pursued. In so
doing we have necessarily assumed that that constitution
is, for our purposes, in a normal state. A true physiology
is not morbid anatomy. We have assumed that from a
study of the structure of man, physical and mental, some
knowledge may be gained not only of his separate organs
21
241
LECTURES ON MORAL SCIENCE.
and faculties, but also of man himself as a system, and of
his end. If he cannot know his own end there can be
no philosophy of man, no comprehension, no satisfactory
knowledge.
That man could, in his present state, know his end without revelation, does not appear. There is no philosophy
in a ruin; and facts are against it. Where the Bible has
not been it does not appear that man has either attained
or retained a knowledge of his true end.
But if it were otherwise, so that the best minds of the
race could reach such knowledge, or even if there were no
moral ruin, yet for a race coming up to moral agency from
the blankness of infancy, and through the long twilight
of youth, it would seem that such knowledge could never
be sufficient as a practical guide. Something more immediate and direct would be needed; and that we find in
those two other principles, Instinct and Faith, in accordance with which it was said in the second lecture that
ends might be pursued.
Having, then, found the harmony there is between the
constitution of man and nature on the one hand, and that
same constitution and revelation on the other; having
shown the relation between virtue and moral good; between individual and general good; between moral and
natural good; between right, utility, prudence, and virtue,
it remains to find the harmony there may be between the
pursuit of ends through these principles of Instinct and
Faith, and by the method already considered.
In comprehending ends man is wholly a philosopher; in
pursuing them he is a practical philosopher. His knowledge becomes power in the highest form, - the power of
attaining his supreme end. But as an end may be attained
242
INSTINCT.
by instinct and by faith, we need to see the relation of a
rational philosophy to the attainment of an end by these.
If we mean by instinct that principle which directs
animals without any comprehension or election of theirs,
which seems, indeed, to be but a higher form of the same
principle that causes the plumule in a plant to tend upwards and the radicle to tend downwards, with no relation to anything higher, then it does not belong to our
subject. But if we mean by instinct that tendency of a
rational nature towards its supreme end which must, as it
seems to us, belong to it if rightly constituted, without
something of which we could not conceive of an end, and
which we may elect to accept or reject as our guide, then
it does come within our range. Then does it become us to
examine both it as a part of our frame-work, and the end
it proposes, and to accept or reject both the end and the
guide. If we accept both, and give ourselves up to the
guidance of the instinct, or the impulse, or the nature, or
whatever we may please to call it, then are we, in an important sense, governed both by instinct and by reason;
and it is obvious that there will be a harmony of the two.
It is the instinct that guides us, but we are not blind in
following it. We trust ourselves to it willingly, as the
muleteer who traverses mountain-passes knows that his
wisdom lies in letting the mule plant his feet where he
pleases. It would, perhaps, be possible for a person to eat
by philosophy. Taking it for his end, an excellent end, to
keep his body of the same weight, he might ascertain by
experiment that there was precisely so much waste of the
system in a given time, and of just such proximate ele.
ments, and he might gather and compound the materials
chemically, and supply them by weight with no regard to
appetite, as he would put so much meal into a bag; but if
243
LECTURES ON MORAL SCIENCE.
he were to do that repeatedly and find the result no better
than when he left the whole business to appetite, and perhaps not as good, it would be rational in him to trust that.
In everything relating to the body, to the preservation of
health, and to its restoration, there is a wisdom of nature
which the wisest regard most, and to which men constantly return after long vagaries of theory, and of what
they call rational methods. Now it may be that what we
call instinct here has not been sufficiently investigated.
We hear men speak of the higher instincts, and of rational
instincts. Are these, then, for the higher nature what the
lower instincts are for the lower? As many view it, what
is conscience but a rational instinct, a guide without comprehension, but rational because it reveals itself as the
voice of God, which all instinct is without thus revealing itself? But if these instincts are the product of the
higher nature, how do they differ from those intuitions
which have been called the product of reason, and so of
the highest form of intelligence? However we may answer these questions, there can be no doubt respecting the
main point of our inquiry.. Whatever there may be of
instinct higher or lower to guide us in the pursuit of a
supreme end, must be perfectly coincident with the impulse to be derived from a rational comprehension of that
end, and in accepting such guidance we may be wholly
rational.
Having thus seen that instinctive morality, if such there
may be, would be in harmony with a rational morality, we
turn to the third mode in which an end may be obtained, that is, by faith, and inquire for the relation of
that to a rational morality. Can a man be rationally governed by faith in precisely the same way as by instinct?
Of the term faith there are different shades of meaning,
244
FAITH DISTINGUISHED FROM KNOWLEDGE.
but its general import is so well fixed that they will give
us no trouble. It is distinguished firom knowledge certain
or uncertain. And, first, fromrn certain knowledge. This
must come directly from the action of some of the senses
or faculties. If we have not faculties that find their evidence in their own activity, we can be sure of nothing, not
even of the being of a God. Our intuitions, those first
truths of reason which are implied in all our other knowing, the legitimate results of the operation of any of the
faculties,- tested as legitimate, the constitution being
given, by their uniformity and necessity, must be received as certainly known. When through these faculties
we have once reached the being of a God, faith in him
would assure us that faculties given by him could not be
mendacious; still, in the last analysis, the evidence of their
trustworthiness must be given in their own activity.
Faith is also distinguished from those beliefs which we
gain from our own processes of reasoning. It is not by
faith that we believe in the result of a mathematical
demonstration, or, if we had never heard of the case, yet
should understand the conditions, that we should believe
the mercury in a barometer would sink if carried to the
top of a high mountain. It is not by faith that we believe
anything that we are required to believe by our constitution or by the laws of evidence, except as confidence in
personal character enters into those laws.
Faith is also to be distinguished from uncertain knowledge.
It is mischievous, as opposing faith to reason, and as
bringing religion into contempt, to make faith something
mystical and obscure, and to which a man may resort
when he is pushed in argument. Says Hamilton, "Faith
belief- is the organ by which we apprehend what is
21*
245
LECTURES ON MORAL SCIENCE.
beyond our knowledge." "In this," he adds, "all divines
and philosophers worthy of the name are found to coincide." Faith an organ! Belief an organ! As if, making
them, as is here done, synonymous, belief were anything
but an opinion not substantiated beyond all doubt. He
might as well make opinion an organ instead of a product.
If all divines and philosophers worthy of the name have
believed this "the more's the pity."
It is also said that faith is that principle of our nature
by which we apprehend the invisible. But what is an apprehension of the invisible but a form of knowledge or
belief based on evidence? If there are principles of our
nature through which we believe in the invisible, they
must be common to all men; but "all men have not faith."
Faith does, indeed, often imply a belief, or, if you please,
an apprehension of the invisible, but that is not its distinctive element. Faith has always a personal element.
It is confidence in a person with reference to anything for
which he offers himself to us. If we believe what a man
says solely because he says it, that is faith. If we believe
it in the face of strong improbabilities from other sources,
the faith is more signal. It is more signal still, if, on the
mere ground of character, and when that stands in conflict
with other sources of belief, we commit to another great
interests. When Alexander the Great drank the cup presented to him by his physician, though he had been warned
by a note that the physician intended to poison him, he
did it by faith. A traveller who should himself know the
way through a forest would walk securely and independently on the ground of his knowledge; but one who
should know nothing of the way, and should commit
himself wholly to a guide, would walk by faith; and if
246
FAITH A NATURAL PRINCIPLE.
his faith were perfect, he would step just as firmly and
securely as the other.
Now, from the condition and circumstances of man it is
plain that faith was intended by God to be a great natural
principle and guide of life. In the absence of instruction
and comprehension, it is to creatures with reason what
instinct is to those without it, and something more. It
was intended to be to them not merely a guide, but a
formative, an assimilative, and an elevating principle. It
is to mere belief what the moral reason is to reason. It
is belief and something more, and is therefore higher. By
the element of belief that is in it, it guides its subject;
and by that which is specifically the confidence, it assimilates and elevates him. It is the one great link, the magnetic link, between parent and child, by which the parent
is enabled to raise the child to his own level. Take this
wholly away, and not only would the improvement of
the race be checked, but improvability could scarcely be
affirmed of it.
That it is a natural principle, is obvious, not only from
its being thus necessary, but because life is full of conditions and relations in which men act from it naturally,
necessarily, and with no feeling of degradation. It is
true universally of children, in their relation to their
parents, and of men generally in their relations to each
other as proficients in specific branches of knowledge, and
that without regard to general superiority. An admiral
may rationally entrust his ship to a common pilot, and a
Newton entrusts his health and life to his physician.
But what is thus natural and necessary in the common relations of life, we might expect, if God be indeed a father, would be carried up into our relations with
247
LECTURES ON MORAL SCIENCE.
hlim, only with such modifications as would be demanded
by his character and those relations. Here, obviously,
reason would demand that the faith should be unwavering
and absolute, stopping at nothing except that which would
make God deny himself. This the very conception of
God as possessed of infinite excellence would require.
From what was said formerly of the identity of a moral
and of the divine law, it will follow that man must be able
to judge to some extent of anything claiming to be a divine
revelation by its intrinsic qualities; and it may be conceded to the advocates of reason that if anything can be
shown to be opposed to the final or highest end of man, it
cannot be from God. HIe cannot require essential wickedness. So much seems to be conceded by the apostle in
the case of Abraham; for he says that Abraham acted on
the supposition that God could reconcile two revelations
which seemed to himt contradictory. "He counted," says
the apostle, "that God was able to raise him up even from
the dead." It is not that, the ultimate end being known,
the insight of reason in regard to that as good can be
shaken, for then would God contradict himself, but that,
in respect to any prescribed means, or to anything short
of the relinquishment of that, faith in God should be
unlimited. At this point it must stop, because the denial
of essential goodness and the denial of God would be the
same thing. If God could command malignity and the
hatred of goodness as such, he would not be God. -
If, then, within this limit, it can be shown that God has
mnade any communication to man respecting his end, - if
he has either told him what that end is, or directed him
how to attain it, - it will be wholly reasonable for him to
receive implicitly what is thus communicated, and to rest
his whole being upon it. Doing thus he is acting upon a
248
I
CHRISTIANITY AND FAITH.
natural and necessary, as well as an ennobling principle,
and his tread may be as firm, and his assurance in regard
to ultimate results as absolute, as if he comprehended the
whole system of the universe from beginning to end and
from centre to circumference. It would not indeed be
philosophy by which he would be guided; but it would
be reason rejecting as inadequate such philosophy as itself
might be able to form, and trusting, instead, to the guidance of Him who is the author and source of all philosophy. It would be the mariner trusting his compass; it
would be the child taking exercise or medicine by the
direction of a father, without knowing the laws of his system. It would, in short, be man understandingly and
rationally taking that place as a child in which is his dignity and his happiness. With sufficient ground of confidence in his father, a child holding his hand -which is
faith- might rationally close his eyes and step where his
father should direct; or, with his eyes open, he might step
in opposition to what would be his individual judgment;
and in these two cases, with the limit above given, we have
the whole relation of reason and faith.
As connected with religion, faith has been the subject of
much discussion, but as a great natural principle of action
it is an illustration of the principle noticed in the first lecture, that what is the most intimate to us, and from the
beginning wholly a matter of course, is the last to attract
attention. When the term was first used in Christianity,
nothing could have been more strange. It was unknown in
philosophy, and it is a strong evidence for Christianity that
it should have thus seized upon a principle which must act
from the first moment of conscious existence, which is in
society what gravitation is among the stars, and without
changing its nature, but only modifying it according to the
249
LECTURES ON MORAL SCIENCE.
relations involved, should have transferred it from its allpervading tlhough unrecognized earthly uses to the higher
uses of religion.
From what has been said it would appear that in pursuing an end from instinct, by faith, or with a full comprehension of both means and ends, we may be acting rationally, while it is only in the last case that we should be
acting philosophically, because a system of philosophical
action can be based only on a conception of ends and of
means. But these three systems or grounds of action,
the instinctive, the religious, and the philosophical, can
have the common characteristic of being rational only on
the condition that they conspire to a common end. That
most clearly they must do. A system not based on the
true end would be erroneous and not philosop)hical; an
instinct or tendency in a being rightly constituted must
prompt to the true end; and faith in God could lead only
to that.
Thus does this philosophy of ends, in connection with
the law of limitation, make provision for the harmonious
operation of every active power in man. In whatever
proportions instinct and faith and philosophy may be
combined, there is yet full provision for the high prerogatives of man as personal and rational, and every power
may conspire to lift him up and bear him on to his true
end.
Having thus brought moral philosophy to a perceived
harmony with those original impulses of the constitution
which are of the nature of instinct, and with faith, which
is distinctively and naturally the religious principle, it will
need but a few words to show its harmony with religion
itself.
It will, first, be a test of any system that may claim to
250
MORAL SCIENCE AND CHRISTIANITY.
come from God, whether it be one of revealed law, or of
a mode of restoration when law has been violated; and,
second, of any such system that should really come from
God it would be the adjuvant.
Moral philosophy analyzes the powers of man, and thus
discovers the true end of each, and so of man himself. If,
then, there be a revealed law, or what claims to be such,
which would require the pursuit of the same end, moral
philosophy must accept that law. It cannot do otherwise.
Then the law is right and binding, whether revealed or
not. If any law claiming to be from God could be shown
to be thus wholly in harmony with the moral constitution
of man, it would be conclusive evidence that it was from
God. It would be a revelation in words of the same will
that had been previously revealed in ends. And this is
precisely what we claim for the Bible as a revelation of
law. What we say is, that no fair and correct analysis of
our faculties can be made that will not necessitate for
them the same end and law that are revealed in the Bible.
So of anything that should claim to be revealed as a
method of restoration. If it could be shown to be not
only in harmony with the law as revealed in the end, but
also to have in it an efficacy so to restore the man that he
shall attain that end, it would be conclusive evidence that
that too was from God. Here the problem would be
double, and the difficulty increased. But as it is the object
or end of the foot that we may walk, and as a rational
physiology would accept whatever would restore a broken
bone so that it should be as good as if it had not been
broken, so, if the moral powers have been injured, a
rational philosophy would accept and welcome any remedial system which it could be shown would enable them to
251
LECTURES ON MORAL SCIENCE.
attain their original end. Here is the test of any system
claiming to be remedial, -harmony with law on the one
hand, and the power of restoration on the other, -and,
tried by this test, we have no hesitation in saying that
Christianity must be received.
252
LECTURE XI.
RIGHTS. - THEIR ORIGIN AND KINDS. - ALIENABLE. - INALIENABLE. -
SLAVERY. - RIGHTS OF PERSONS AND OF THINGS. - GIVING AND RE CEIVING. - RIGHTS OF GOVERNMENT.- LIBERTY AS RELATED TO RIGHTS.
-DIFFERENT KINDS OF LIBERTY-NATURAL, CIVIL, POLITICAL.
OF any correct system of moral philosophy one characteristic must be that the active powers will, in their movements, harmonize with each other. That they do this
in connection with the system of ends, we have seen.
Through the law of limitation each higher power is harmonized with the lower, while the highest is left to act
fireely and to expand in its connection with those infinities
to which it is naturally related. This gives us a philosophical system for the individual which we may comprehend.
But not only may we comprehend both means and
ends, and so seek them intelligently; we may also seek
ends from a native tendency involving in it, if it be not
instinct, the instinctive principle; and we may seek ends
by faith. These principles may be combined in very diffierent proportions. They must be, as persons are younger or
more advanced, as they are ignorant or instructed; but it
was one object of the last lecture to show that whatever
the proportions might be, these principles might be so
accepted and permeated by the rational nature that we
should be rational in acting from them, and that they
would be in perfect harmony.
But God does not regard the individual only. He has
22 253
LECTURES ON MORAL SCIENCE.
instituted families, communities, nations. Of these he designed the well-being, and has provided for it in the organization of man. Would, then, this doctrine of ends, with
its law of limitation, be an adequate basis for social order?
As an individual, man is to do right; as a member of a
community he has rights. What it is on this system to
do right, we have seen. Would there also grow from it a
perfect system of rights? If so, there would be in it an
adequate basis of social order, because of that it is the
one condition that every man shall have his rights. If so,
we may well accept a doctrine thus providing for the right
ordering not only of the individual, but Of the community.
On this system, we have seen that that is right which a
man must do that he may attain the end for which God
made him. Rights must, therefore, be based on the relation of those things to which we have a right to the attainment of our own end or that of others. A man will have
a right to everything that is essential to the attainment of
the end for which he was made. So a parent will have a
right to everything which is essential to the attainment of
the end for which God made him a parent; and society
and government will have a right to everything necessary
for the accomplishment of the ends for which they were
instituted -just that, and no more.
An exclusive capacity, inherent or given in the order of
nature, together with a disposition to confer upon others
what is essential to their end, is the ground of rights over
them. Hence the rights of God, of parents, and of governments. A necessity for anything essential to his end is
the ground of a claim by the individual upon any who, in
the order of nature or of providence, may have the exclusive power to meet that necessity. Hlence the claims of
children, of citizens, of the poor, of humanity.
254
ORIGIN OF RIGHTS.
We have here the general principle; and if it be correct, then will the basis of right and of rights be the
same, only it will be viewed in different relations. We
shall have, moreover, what is not a little desirable, in the
distinction drawn between the higher and lower powers, a
measure of rights as more or less important and sacred.
Thus we shall have rights from the instincts,-that is, those
which would respect the attainment by instinct of its end;
and rights of the appetites, or those which would respect
the attainment by them of their end; and so of the desires, and of the intellect, and of the natural affections,
and of the moral and spiritual nature. Certainly we may
say that he who should be in no way so encroached upon
or obstructed that he should be unable to attain in the
best way all the ends indicated by these different active
principles might be said to have all his rights; and if he
welre so encroached upon that he could not reach perfectly
any one of these ends, he would not have all his rights.
The truth seems to be that in the tendency of every active
principle towards its end there is the voice of God; and
that when, through the intervention of others, there is an
obstruction to the attainment of its ends, that voice utters
itself through the moral nature in the assertion of rights.
That this is the history of the idea and sentiment of
rights -for it is not merely a sentiment -seems probable,
because it is foreshadowed by what occurs among animals.
That they have the perception of relations and the sentiment that we have, cannot be supposed, but practically
they assert what seem to be rights, and what is analogous
to them, on the same principle. Let an animal have an
instinct, or an appetite, or a natural affection, as that of
the parent for its.offspring, and it will be found that it will
be ready to resist and beat off all intrusion that would,pre
255
LECTURES ON MORAL SCIENCE.
vent it from accomplishing the end thus indicated, and the
strength of endeavor will be proportioned to the importance of the end. So is it with man. He is prompted
by some original impulse to the attainment of an end.
This would imply struggle against obstacles, and the resistance of any interference that would prevent the attainment of the end. It is in connection with such promptings
and resistance that the moral reason necessarily forms the
notion of rights, and that the sentiment is felt; and thus
that which with the brute is defended simply by force,
comes with man to be guarded by the most sacred sentiments, and to be fortified by laws, and customs, and institutions.
From this view of the origin of rights it will appear
that the idea of right is the primary, and that of rights
the subordinate and secondary idea. A man has rights in
order that he may do right. If there were no end, and
so nothing right, there could be no such thing as rights.
Hence rights, however real and important, may never be
defended at the expense of right. A man may be deprived of all his rights, but he may not cease to adhere to
that which is right.
At this point it is that we may see how it is that the
destiny of a man, that is, his highest and ultimate destiny,
can never be taken out of his hands. Men may deprive
him of every right, but they can bring about no combination of circumstances under which it will be impossible for
a man in those circumstances to do right. It may be a
fearful alternative, and there may be unspeakable wickedness in presenting it, when a man mqt be deprived of his
rights, even of that to life, or cease to do right, but it is
the glory of man's nature that there is in it the capacity
of adhering to what is right under all deprivation and
256
RIGHTS - ALIENABLE AND INALIENABLE.
all suffering. If it were not for this - the higher estimation of right than of rights - no man could be a martyr.
Right belongs to man in his individual capacity, rights
from his relation to others.
Of rights as thus originating and thus distinguished
from right, some are alienable, and some inalienable; and
we find in the distinctions already laid down the ground of
this difference. An inalienable right is one which arises
in connection with the pursuit of our highest end. With
that nothing may interfere; and a right thus based is called
inalienable because it cannot be parted with freely without
crime, and cannot be rightly taken away unless forfeited
by crime.
As has been seen, the moral, no less than the physical
nature, has its end; in the use of means for the attainment
by that nature of its end, the idea of rights the most
sacred would arise; and to whatever is an essential condition for the attainment of that end man has an inalienable
right. With that he may not consent to part, and no one
may rightfully wrest it from him; but any right which is
not thus necessary he may alienate.
After the moral nature, the natural affections and the
intellect are next in dignity. That the rights which originate in connection with the exercise of the affections are
alienable, appears, since a parent may transfer to another
all the rights and responsibilities vested in him as a parent.
A child may be wholly given away,.its name changed, and
the rights of the parents over it vacated according to law.
Than this, perhaps a stronger case could not be put under
the rights of the affections. Of the intellect it is to be
said that its operations are so essential to the full attainment of the ends of the moral nature that it can hardly
stand on its own ground; but that a man may employ his
22*
257
LECTURES ON MORAL SCIENCE.
intellect for gain at the will of another, and, so far as that
is possible, wholly give it up to his control, provided that
control does not interfere with the attainment by the
moral nature of its end, will, I suppose, be conceded.
Why not? The intellect is simply instrumental, and may
be employed by the executive power in any way that shall
not contravene a moral end. Of the rights that originate
from the desires, as that of property, I need not speak, as
it is conceded that they are alienable.
All inalienable rights may be included in those of life
and liberty. A right to the pursuit of happiness, mentioned in the Declaration of Independence, would be
included in that to liberty, since no man can have liberty
who is debarred from the pursuit of that. And yet liberty is not wholly inalienable. In some respects, and to
some extent, a man may part with his liberty, and it has
not always been easy to say how far he may go in this,
According to the principles already stated he may part
with his liberty in any respect, and up to any point, that
shall not interfere with the attainment of his highest
end. Beyond this he can make no contract that would
not be unlawful, and so not binding; for man has a paramount duty to God respecting himself, which is as fully
binding as any other duty. He may never lawfully do
anything with himself which shall prevent the great purpose that God had in view in giving him being from being
accomplished. Except as an indispensable condition for
a higher end, there is nothing sacred about liberty; it
is capable of being wholly abused, and if it may be conceived that a higher end may be promoted by giving it
up, then may it be given up. According to this those Moravian missionaries who sold themselves into slavery that
258
SLAVERY.
they might preach the gospel to the slaves, may have been
justifiable. They sacrificed liberty for a higher end.
Inalienable rights are those of which a man cannot
divest himself by contract; which he may not, under any
circumstances, lawfully demit; but he may forfeit them by
crime, and be wrongfully deprived of them by others. It
is in this last case, in the violation of an inalienable right,
that the greatest wrong is committed, and of this we see
the reason in what has been said respecting the ground of
inalienable rights. To deprive a man of life is everywhere
regarded as the highest crime; and next to that, in some
circumstances perhaps even greater, is the crime of depriving him of his liberty. When this is so done as to
degrade a human being, and to come between him and his
highest end, we have a crime that involves in it the essence
of all crimes.
Of slavery, so far as it interferes with inalienable rights,
our abhorrence cannot be too strong. It interferes with
other rights, as those of the desires. It takes property, or
the labor that makes property, without an adequate compensation. It violates the rights of the affections. It sep arates husbands and wives, and parents and children;
putting, in the eyes of the law, and often practically, and
by the necessities of the system, the natural affections of
the slave on the same level with a brute instinct. It inter feres with the rights of the intellect. It keeps men in
ignorance, and prohibits them from learning to read the
word of God. It gives the slave no security for anything.
Everything must depend upon the will of the master; and
if that will be reasonable, then upon his life. Now, while
it is true, as has been said, that no man can be so placed
that he cannot adhere to the right, yet such a system,
applied to masses of human beings, must degrade them,
259
LECTURES ON MORAL SCIENCE.
must come between them and their highest good, and so
touch inalienable rights. The highest right of a man is his
right to himself, and any right of property that would so
contravene this that man shall be treated in any way as a
brute, or degraded, and that would come between him and
his end as designed by God, is impossible. No man can
give it; the man himself cannot; no state can give it, and
any attempt to hold such property is sin per se.
That there may be a temporary and modified system of
involuntary servitude without infringing upon inalienable
rights, and with ultimate benefit to those so held; that
under a system of perpetual servitude the actual guilt
will depend much on the light of the master, and the
spirit in which it is administered; and that, under peculiar circumstances, the legal relation of master may be
sustained for the good of the slave, not only without
guilt, but meritoriously, may be conceded. And it is
because this partial alienation of liberty without degradation is possible; and because guilt is so modified by
acquiescence in established customs to which men have
been used from their infancy, and which they have been
taught are right; and because, from obstacles to emancipation through wicked laws and the disabilities they lay
upon the freed-man, or from the helplessness of infancy
or of old age, the legal relation of master may sometimes
be rightly held while yet the system itself is one of utter
oppression and wrong, often and generally infringing upon
inalienable rights; and because of the immense pecuniary
interests at stake, that it is possible for men to hold such
discordant views on this subject, and that their views
are held in connection with feelings so intense.
Having thus seen what is the origin of rights, and the
distinction between those that are alienable and those that
260
RIGHTS OF THINGS AND OF PERSONS.
are inalienable, we turn to another distinction. There are
rights which have it for their object to guard the individual
against the encroachments of others. As thus used, the
sole correlative of rights is obligation, and it is in this aspect that rights are more generally treated. If I have a
right to a piece of property, all others are under obligation
to abstain from its use. The object of such rights is so to
protect the individual in his freedom, that he may accomplish the ends indicated by his active powers. Such rights
respect things, and not persons; or, if they respect persons,
it is only as they are so related to us that we may by them
accomplish our own ends.
But there are also rights over persons. The object of
these is to enable those in whom they are vested to aid
others in the accomplishment of their ends. Here the
correlative of rights is still obligation. If the parent has
a right over the child, the child is under obligation to
respect that right. But here the right involves by necessity not only an obligation on the part of others, but also
a duty on the part of him in whom the right is vested,
and this duty thus necessarily involved in the right, and
measured by it, may also properly be called its correlative.
The foundation of the right here and in the other case is
radically the same, as they both have reference to the
attainment of an end; and yet there is an essential difference. In the first case the ground of the right is the necessity of that to which the individual has a right in order
to the attainment by himself of his own ends, that is, of
those indicated by his various active powers. But in the
relations of society human beings are not always capable
of attaining these ends without aid from others. In that
case others may have rights over them, natural or acquired; but the ground and measure of those rights will
261
LECTURES ON MORAL SCIENCE.
be found in the necessity there is for aid in the accomplishment of those ends, and in the power and duty of
those who possess the rights to render that aid.
In what has now been said we have a clear distinction
between rights over things and those over persons. This
distinction was indicated by Blackstone, under the heads
"Rights of Things" and "Rights of Persons," but his statement of any ground for it is so indistinct that in a note to
Chitty's edition of his work it is said that "the distinction
of rights of persons and rights of things in the first two
books of the Commentaries seems to have no other difference than the antithesis of the expression." As the most
he could make of it, the annotator adds, "The distinction
intended by the learned judge in the first two books appears to be in a great degree that of the rights of persons
in public stations, and the rights of persons in private stations." This is wholly aside from the real ground of the
distinction. As has been said, rights over persons have
respect to the accomplishment of ends by those persons,
and involve duties; while rights over things respect the
accomplishment of ends by ourselves, and do not in the
same way involve duties.
This distinction is needed because the rights over persons are numerous and important, and without it we have
no way of fixing precisely the ground and limits of those
rights. These are the rights of parents, of guardians, of
teachers, so far as they have also guidance and control;
they are, in general, the rights of those that govern; and
have, standing over against them, not only corresponding obligations, but also corresponding rights. Wherever
there is a right to govern, there is a corresponding right to
be governed rightly. What it is to be governed rightly is
implied in what has already been said. A man ought to
'"' 6 2
NATURAL LIMIT OF RIGHTS.
govern another on the same principle on which he ought
to govern each of his separate faculties, and his whole
self. He governs those faculties rightly when he causes
each to accomplish its end. He governs himself rightly
when he accomplishes his own end; and he governs another
rightly when his government is wisely directed to enable
that other to accomplish his end. This is the law of limitation here. Hence the parent has a right, so far as the
destiny of the child is committed to him, to all the control
necessary to secure for the child its true end. Whatever
power he may use for any other end is not properly that
of a parent, since it would not grow out of the parental
relation as instituted by God. That relation is one of
guardianship of the child with reference to the ends for
which he was made, and especially to his highest end;
and if the child could certainly know that he could secure
his highest end only by disobeying his parent, he would
be bound to disobey him. This shows the natural limit to
the rights, and so to the authority of the parent. And
what is thus true of a parent is equally true of a guardian,
a teacher, a magistrate, a government. So God governs.
This is the model he sets before us, and he has given no
rights to any of his creatures that will justify them in
governing upon any other principle.
Very beautiful is the relation thus established between
the governing and the governed, and quite in accordance
with what has been previously said. We have seen how
beautiful is that relation of all things as conditioning and
conditioned by which there is a continual subserviency of
that which is lower to a higher end, till this universe, as
more immediately known to us, is built up from its base to
its apex, and culminates in man. In this process the lower
force is independent of that which is higher and unmodi
263
LECTURES ON MORAL SCIENCE.
fled by it till we reach organization. In all arganization,
while the higher is built up by the lower, and constantly
sustained by it, yet the higher reacts upon the lower, and
becomes in its turn essential to that. The stomach and
digestive system are for the brain. They build it up; but
the brain reacts upon them, and unless it be healthy they
will fail. Where organization begins, the movement within
each organism becomes circular, and not merely one of upbuilding from a base. And now, when we pass into the
region of intelligence, we find provision not merely for a
system of forces acting from below to build up that which
is above, but that there shall be forces from above intelligently acting to benefit if not to elevate that which is
below. At first, and in mere organizations, the lower
builds up the higher, and sustains it, and is wholly for
that. Any action from the higher to the lower is simply
to sustain the lower in its own place and function as tributary, but never to elevate it out of that sphere. But when
we reach the sphere of intelligence the object of the action
from above is to elevate the lower. When the summit is
reached, then, through this arrangement of rights and of
duties, a circle is formed by which the system works from
the top, so that that which is spiritual is drawn up from
above, since there could have been no force from below
adequate to push it up. Certainly the parent, as a parent,
is for the sake of the child, and his end in that relation is
accomplished when he has brought the child up to his
own elevation, or, rather, to what that elevation ought to
be. In doing this there may be, there ought to be ties
formed that shall be permanent, that, as spiritual, shall be
eternal, and so the highest here minister to that which is
still higher; but the parental office, and the merely natural
affections connected with it, have exhausted themselves
264
RECEIVING AND GIVING.
when the parent who is what he should be has raised the
child to his own elevation. So all analogy teaches. So
is it with every animal that has natural affection; and
where provision is made for the young independently of
the parent, the affection is not given.
Up to the point where giving from above begins to elevate that which is below, if there may be said to be blessedness at all, it had been more blessed to receive than to
give. The giving was always by the lower to the higher,
and for the sake of the higher. But we now reach a point
where the giving is by that which is above, for the sake of
that which is below, and God has connected with it, in
the end which it accomplishes, in the affections which it
gratifies, and in the improvement, and growth, and dignity
of the giver, a blessedness that could not come from receiving. It is here, indeed, that we have the element of
the noblest giving. It is not simply that which addresses
itself to the animal nature and satisfies want. It is that,
when it is needed, always that; but enwrapping and bearing with it a giving of affection and self-sacrifice that
would lift up that which is below it; and if this element
be wanting, no giving can avail much, and the highest
blessedness of it cannot be known.
In speaking of the rights that involve duties, I have
referred almost wholly to those of parents; but the principle applies equally to the rights and duties of society
and of government. These, scarcely less than the parental
relation, are essential for the perfection of the individual.
Without them he cannot find his sphere, and scope for the
expression of his whole nature. It is, indeed, that constitution by which society is thus necessary, that makes a
number of individuals a community, that makes the state
an institution of God, and the race a unity. An exclusive
23
265
LECTURES ON MORAL SCIENCE.
capacity to confer such aid, given in the order of nature,
confers rights. A necessity that such aid should be conferred is the ground of a claim. Hlence the reciprocal rights
and duties of the famnily, of the state, and of humanity.
It is to governments as founded upon this principle that
attention is especially needed. Practically, they have too
often been instruments of oppression. They have kept
down and degraded the governed. It is among the saddest features of the history of our world that the very
conception and ground of this beautiful and beneficent
function of government should have been so wholly lost
sight of, and government so perverted to purposes directly opposite to those for which it was intended. A vast
abstraction, or, if you please, a general conception called
the state, has been idolized. It has been supposed that
the individual was wholly for that; and so, partly through
a blind and perverted instinct of patriotism in the people,
the very institution which ought to have been the most
efficient for their elevation has often been the most potent
engine for their oppression and degradation. So it has
been; so it is still.
But in the light of our discussion, government has no
right to be, except as it is necessary to secure the ends of
the individual in his social capacity; and it must, therefore,
be bound so to be as to secure these ends in the best manner. This is the whole principle, and only the full application of it is needed to make governmental and social
movements on the earth correspond in their order and
beauty to the movements of the heavens. On this principle there could be no conflicting rights as between the
individual and the government. The government could
require -from the very ground of its rights as already
stated, it could have a right to require - nothing that
266
RIGHTS OF GOVERNMENT.
would not be in harmony with the ends of the individual,
and whatever the government might need to accomplish
its ends that should be thus in harmony, the individual
would be bound to concede.
This is the general statement. To obviate practical difficulties, however, it must be observed that when it is
said that government is for the individual, it is not meant
that it is for any one individual especially, but for all the
individuals of whom the society is composed. If, therefore, a case should occur in which the good of the society
would require that the alienable rights of the individual
should be, not, as is generally said, given up, but alienated
for an equitable consideration, the government, as the agent
of society, has the right to enforce such alienation. This
paramount right government has, and must have, firom the
end for which it was instituted. Is it for the good of society that it should take the land of a man for a road? It
has a paramount right, and takes it; but it gives him an
equivalent. It would be thought monstrous to take it
otherwise. Is it again for the good of society that it
should take the time, more valuable it may be than land,
of an innocent but accused man, that he may be tried? It
has a paramount right, and takes it, but it makes no compensation. But that society.is bound in equity to make it,
there can be no doubt.
The right of society is to take, for its own good, the
alienable rights of the individual, on condition that those
rights shall be surrendered only for a fair equivalent. This
the ends of society, and so of government, require. What
society says to the individual is, "We will give you as
amrnple means as you now have to accomplish your ends; we
interfere therefore with none of your fundamental rights;
but we cannot suffer mere will or caprice to stand in the
267
LECTURES ON MORAL SCIENCE.
way of the good of the whole." The right to say this I
do not suppose society gets from any consent of the individual, or any agreement on his part to surrender certain
lights; but because it is of divine origin, and has, therefore, an inherent right to accomplish its ends. This it
must do for the sake of the individual, since his perfection
can be reached only through society, and hence, according
to the doctrine stated, the individual can have no rights not
compatible with the ends of society. All alienable rights
must be held by him subject to the condition that when
they interfere with the good of the whole, they shall so far
cease to be rights that they may be alienated by the will
of society regularly expressed through its government, and
for a fair compensation. According to this view, the rights
of the individual and of society would be perfectly harmonized. At least, there could be no conflict in regard to
alienable rights. Nor could there be any respecting those
that are inalienable, since those are sacred. Those society
may not touch. It is impossible that any legitimate end of
society should be gained by trenching in any degree upon
any inalienable right, and therefore society can have no
right to do so. Injustice, tyranny, may do anything. A
triumph of wrong there may be, but there can be no conflict of rights. 1
From the consideration of rights we pass to that of liberty, of which the conception of rights is both the basis
and the natural limit. Rights and liberty! These are
among the most exciting and stimulating words of the
English language, and unless our view of their grounds
and limitations be distinct they may become words of delusion and mischief, - cabalistic words for the popular
declaimer and demagogue to conjure with. It is a slow
process by which the conceptions connected with such
''6 8
LIBERTY AND LAW.
words in the popular mind become clear and steady; but
nothing is more needed, especially in a government like
ours; and whoever contributes to it in any degree is doing
the public good service.
It has just been said that rights are the basis and natural limit of liberty. If there were no rights there could be
no law. God would have no right to give or to enforce
one, and there would be nothing for law to guard. Law
is the guardian of rights and the condition of liberty, since
without law there would be anarchy, which is the opposite
of liberty. It is, I know, usually thought that the idea
of liberty is the primary one, and that of law secondary,
as coming in to restrain liberty; but if we take law in its
widest sense as that which gives stability and regularity,
and a rational ground of expectation, we shall see that
without the conception of that there could be no ground
of choice or of action, and so none for liberty; so that it
may well be doubted whether the conception of law does
not underlie that of liberty of any kind, as it certainly
does that of all desirable and rational liberty. We may,
indeed, conceive of what is called absolute liberty, by
which is meant a liberty of doing, without question or
control, whatever the individual pleases. For a single isolated individual this is conceivable, but not in a community of individuals, each having free will and independent choices. Such a liberty would be an element of utter
confusion, like that which would ensue in the physical elements if their affinities were unloosed and wholly capricious, so that there were nothing of regularity in their
movements. We may well conclude, then, that the first
liberty among created beings was born and cradled and
trained amidst the sanctities of law, and that any exercise
23*
269
LECTURES ON MORAL SCIENCE.
of it except under the control and guidance of law must
be a curse.
Taking liberty, then, as known by us, and desirable for
us, we inquire after its different kinds.
And, first, there is natural liberty. This is not absolute
liberty. That is not natural. The law of any being is
indicated in its nature, and no liberty can be natural that
would overstep that natural law. Than this nothing could
be more unnatural. But natural liberty is simply that
which is commensurate with natural rights. Every man
has originally a natural right to use all the means furnished him by God for the attainment of the legitimate
end indicated by each of the active principles of his nature; and his natural liberty would be such a freedom
from restraint that he could avail himself of all such
means for the attainment of those ends. If God has not
furnished the means of doing this without encroaching
on the rights of others, then his liberty and his rights find
their limit together. Let a man have a liberty by which
he may attain every end of his being which God has given
him the means of attaining; let no man come wrongfully
between him and those means, and he has all the liberty
that any being ought to have or that can be natural to
any. This, then, is natural liberty- a liberty to use all
the means that God gives a man for the attainment of the
ends indicated through his nature.
A man has natural liberty to use the means above indicated. Has he also a natural right to defend himself in
such use? This is commonly said, and that the right of
society to defend the individual in the use and enjoyment
of these means is from a voluntary transference of such
right by the individual to society. But society is natural,
and as such has rights and duties; and looking at its end,
270
CIVIL LIBERTY.
we shall find it to be its natural right and duty to protect
the individual in the use of these means. If, in an unnatural state of isolation, the individual has a right to defend
himself in the enjoyment of these means, it may be quite as
correct to say that he gets that right by a transference of
it from society, as to say that, in ordinary circumstances
and in a natural state, society gets the right by a transference of it from the individual. Saying this we avoid all
conflict of rights; we avoid the unnatural and violent supposition that man is necessitated to give up any of his
natural rights in order to secure the remainder.
We next inquire what civil liberty is. Of this it is
thought by some that the notion is so complex that it cannot be defined, but only described, and the circumstances
stated in which it is enjoyed. But having seen what natural liberty is, we say that civil liberty is natural liberty
under the guardianship and guarantee of an organized
society. It is the liberty which a man enjoys when his
rights are protected and guarantied by society instead of
himself. Hence the only abridgment, if such it may be
called, of natural liberty needed that it may become civil
liberty will respect the means of attaining the end, and
not the end itself. Man has a natural right both to defend
and to use and enjoy the property produced by his own
labor; but the right of defence vests in society, and by
suffering it to remain there he enjoys greater freedom and
security in the use of his property. This freedom and
security are civil liberty. It consists in a liberty and security in enjoying the end which can be attained only by
leaving with society the responsibility, and so the right of
protecting ourselves in that enjoyment. But protection is
not in itself an end; it is only a means; and therefore we
are always to remember that in requiring us to leave with
271
LECTURES ON MORAL SCIENCE.
society a portion of our rights civil liberty only requires
us thus to leave the right to employ means, but never to
abandon the right to enjoy ends.
The perfection of civil liberty will be measured by the
degree of freedom and security that can be attained in the
enjoyment of every right which can be left to the guardianship of society. Every right cannot be thus left, and
the liberty to defend such right must remain with the
individual; but perfect civil liberty will be the greatest
possible freedom and security under the guardianship of
society for every right that naturally belongs to its care.
It may be added that, under civil society, such liberty, as
essential to the end of the individual as a social being, is
a natural right.
In the view of it above taken civil liberty is for the perfection of the individual. It may also be regarded as it
stands related to the ends of society as a whole. The
object is not only to find the point where the action of the
whole shall either be for the good of the individual, or not
militate against it; but also where the action of the individual shall either be for the good of the whole, or not
militate against that. That these points may be found,
and that they would coincide, cannot be doubted by any
who know the balancings, and adjustments, and harmonies
there are in the works of God, and for which we might
expect the most perfect provision in the highest department of those works. This would bring the interests of
the individual as an individual and his interests as a member of the community into perfect harmony. In this point
of view civil liberty would be conditioned on such a restraint of individual action as should guard the interests
of the whole from injury.
Civil liberty, as has been said, is a natural right. Hence,
272
THE RIGHT TO VOTE.
under a free government, it is accorded to all, whether they
are citizens or not. But the right to take a part in the
government, to say who shall administer it, and what provisions shall be made for the maintenance of civil liberty,
is not a natural right. It belongs to society as a whole,
but not to every individual in that society. It is not generally supposed to belong to minors, to women, to foreigners, except on specified conditions, and in the most of our
States it is not granted to the free blacks. What the principles are on which this right should be conceded, and
what should be their application in particular cases, it is
not always easy to say. Here an end is to be secured.
Society is bound to secure it in the best way it can, but
the means and materials for doing this may be very different at different times and in different nations.
On this point it may be said, first, that a reasonable presumption of hostility to the welfare of the society would
be a sufficient ground for excluding any one from having
a voice in the government. Hence criminals are excluded;
and there may be factions, or races, known to be hostile to
the government, who may be justly excluded while that
hostility remains.
Secondly. Incompetency to understand and promote the
ends of society would be a sufficient ground for exclusion
from political rights. It is on this ground that minors are
excluded, and foreigners who are presumed to be ignorant
of the nature and working of institutions under which
they have but recently come. It is true that many minors,
and many foreigners not naturalized, are better qualified
to exercise political rights, and so for what is sometimes
called political liberty, than many who do exercise those
rights; but where there is no absolute right society may,
and must, fix the best average limit it can. According to
273
a
LECTURES ON MORAL SCIENCE.
this, under institutions like ours, society would have a right
to say, as has been proposed, that no man should vote who
could not read. It may be expedient in given circumstances that such persons should vote, but they have no right.
It may be wrong that they should be permitted to do it.
Society cannot be bound to entrust its interests and destinies to ignorance, or chance, or passion.
Once more; if there be such relations established by God
that one portion of the community cannot take part in administering the government without injury to the ends of
society, then that portion may be excluded. It must be on
this ground, if upon any, that women are to be excluded
from the right of voting and holding office under our government. They cannot be excluded on the ground that
they are not interested in the welfare of the government,
or that they are incompetent. But it is never safe to violate any true instinct of humanity. There are some things
that depend not so much upon reasoning as upon sentiment and a felt propriety. When a country is invaded
and civil liberty is to be defended, it is not so much from
any laying down of principles and formal reasoning as from
a felt propriety that the women remain at home, while the
men go to the battle. In the same way, when civil liberty
is to be instituted and sustained, it may be from the same
felt propriety that men alone should be concerned in the
conflicts of public debate, and at the polls. It may be
that in her relations to man, when she is elevated to her
true position, God has made provision that her influence
shall as effectually reach a free government for good as if
she were immediately concerned in it; or if not, there
may be obstacles which would render it inexpedient that
she should have that power at present; and in either case
society would have the right to withhold it. Certainly, if
274
DUTIES INVOLVED IN RIGHTS.
there be such relations established by God that one portion of the community cannot take part in the government without injury to society, then that portion may be
excluded. How far this may be the case in any particular
instance, each society must judge for itself, as it does upon
other and similar questions. I
I cannot close this lecture without observing that this
subject of rights, regarded as a barrier against encroachment, and as involving duties, demands the especial attention of a free people. Among such a people there will
always be a tendency to regard liberty as a right of unrestrained action, and rights as something to be enforced.
It is those days when liberty was gained and rights enforced that nations celebrate. But is easier to gain liberty
and enforce rights than, having gained them, to practise
the self-control that shall respect rights, and the self-denial
and faithfulness and patient waiting required in performing the duties that our rights involve. This is the turning
point with us. Can we use our freedom and enjoy our
rights without encroaching upon the liberty and the rights
of others? Will parents, and magistrates, and citizens,
fulfil the duties that correspond to their rights? Will
they see that individual and unauthorized action is so
restrained that all shall have their rights? There is no
grander sight than that of a great people, powerful and
free, under the guidance of a comprehensive wisdom,
always arresting its action at the point where it touches
the rights of others, protecting those of the most feeble,
and trusting calmly for its aggrandizement to the gradual
but resistless power of intelligence, industry, and freedom,
under the guidance of justice. And there is no sadder
sight than such a people governed by fraud and cunning,
torn by faction, disintegrated by selfishness, denying to
275
LECTURES ON MORAL SCIENCE.
others what they claim for themselves, with no faith in
the natural power of free institutions to perpetuate and
extend themselves without force, and thus putting into the
hands of others a cup, which, in the circuit and balance of
God's retributions, must be returned to their own lips, and
which they must be Cbmpelled to drain to the very dregs.
276
LECTURE XII.
A FUTURE LIFE. -ITS RELATION TO MORALITY.-THE PHYSICAL ARGU MENT. -MORAL ARGUMENTS.
WHAT man ought to do will depend on the end for
which he was made. If he was made for this world only,
then he ought to live for this world. But if he was also
made for a life after this, and his conduct in this life would
affect his condition in that, then he ought to live with
reference to that. We labor for the morrow, because we
expect to awake in the morning. It is thus that the doctrine of a future life connects itself with morality; and as
we have seen that man is connected with all that is
below him, it will be a fitting close of our subject to
inquire what indications there are in his nature that he is
also connected with that which is beyond and above him.
Than this no inquiry can be of greater interest.
Whether there is a God or not; whether this visible structure of the universe is to be eternal or not; whether the
generations of men are to be perpetuated, or are to be destroyed by some general convulsion of nature, are questions
that little concern the individual man if he is evoked into
being like the bubble upon the ocean, to appear but for a
moment, and then vanish forever.
The first indication of a future life that I shall mention
is drawn from the nature of the mind as simple and indivisible, and so incapable of destruction except by annihilation.
277
24
LECTURES ON MORAL SCIENCE.
Concerning that which underlies the power of thought
three suppositions may be made, and only three. It must
belong either to one single, indivisible, ultimate particle of
matter; or to a number of such particles united together;
or to what we must call an immaterial substance entirely
distinct from matter.
Does the power of thought, then, reside in a single, indivisible, ultimate particle of matter? I think not, because
these particles are so minute. No microscope can reach
them. If a single grain of the salts of iron be put into
thirty thousand pints of water, it can be detected by experiment in every drop of that water. A hare, in his flight,
leaves particles of insensible perspiration upon the earth at
every footfall. These must be inconceivably minute, as
they are constantly given off so long as the hound can foltow the track. But to suppose that one such ultimate particle has, in addition to the properties of matter, those of
thought, feeling, memory, imagination, judgment, that it
studies fluxions and metaphysics, indites poems, and governs nations, seems absurd.
But I need not dwell on this, because those materialists
who deny a future life do not advocate it, and for the very
good reason that it would be a strong argument against
them. If the soul be such an ultimate particle, then it can
perish only by annihilation, and it seems to be a principle
in the government of God not to annihilate anything.
What we call destruction is simply a change of form,
never an annihilation of substance.
Is, then, the power of thought the property of a number
of particles of matter united together?
Here again we must look at the constitution of matter.
Concerning this there are two suppositions. One is that of
Boscovich, and was adopted by Priestly, a distinguished
278
THOUGHT SIMPLE.
materialist. The supposition is that what we call matter
consists, not of solid particles, but of centres of attraction
and repulsion. As other philosophers have said, take away
solidity and matter vanishes, so Priestly says expressly,
"Take away attraction and repulsion and matter vanishes."
This seems to me to deny the existence of matter as a substance, though not as a force, and it cannot be necessary
in opposing materialism to show that thought cannot be
the property of a number of centres of attraction and repulsion, when, by the supposition, those centres themselves,
as material bodies, do not exist.
We take next the common supposition that matter consists of solid extended particles of great minuteness.
Whether such particles are ever so united that there is
actual contact between them is not decided; but whether
there is? or not, we must remember they are separate and
independent bodies, and that a body which we call one is
not a unit, but a collection of units to which we give a
common name. There is no unity till we come to ultimate
particles, or to mind.
Now the supposition is that thought, though not the
property of any one of these particles separately, is yet the
property of a number of them, greater or less, united together.
But this is surely contradicted by the consciousness of
every man in regard to the oneness of that being which he
calls himself. It is also contradicted by the nature of the
mental phenomena, as thought, feeling, consciousness,
which are simple, and incapable of division. If this doctrine be true, then the thought, originating not solely in
one particle, but in a number, must come, part of it
firom one, and part from another, and what is thus made
up by composition may be again divided. According to
279
LECTURES ON MORAL SCIENCE.
this there would, as has been said, be no impropriety in
speaking of the half or the eighth of a thought, of the top
and bottom of a feeling, of the east and west end of con
sciousness.
But, again, if this doctrine were true, there could not
only be no such thing as simple indivisible thought, but
there could be no personal identity. Our bodies undergo
constant change; they are no more the same bodies for
two days together than the stream which we pass over on
two successive days is the same water. The brain participates in these changes. I remember now what happened
when I was four years old; but there is not in my system
now one particle of matter that was there then. How,
then, does this new matter know what happened to the old?
How can this consciousness, this sense of identity, be transferred from one particle to another? According to this,
we should be undergoing a continual death, for, as the
whole brain dies when it ceases to think, so there must be
some particles of it, as they are passing off, constantly giving up the ghost, and leaving their transitory honors to
their successors. And these others, -how are they exalted! That which was yesterday a portion of a potato or
of a calf's brains, may to-day become a part of the soul of
a philosopher! That there are any who believe this is
the most plausible argument that I know that their souls
are thus made.
In reply to this objection I have never seen anything
better than the following ironical answer from Martinus
Scriblerus: "Sir John Cutler had a pair of black worsted
stockings which his maid darned so often with silk that
they became at last a pair of silk stockings. Now, supposing those stockings of Sir John's endued with some
degree of consciousness at every particular darning, they
280
MATERIALISM.
would have been sensible that they were the samne individual pair of stockings both before and after the darning,
and this sensation would have continued in them through
all the succession of darnings, and yet, after the last of all,
there was not, perhaps, one thread left of the first pair of
stockings, but they had grown to be silk stockings, as was
said before."
But however conclusive the above arguments may seem,
I am aware that I have not yet touched the real difficulty
as it lies in your minds, if you have been accustomed to
read a particular class of writings on this subject. It is,
that thought is never manifested except in connection
with a brain or nervous system, on which it seems to
depend; that as one changes the other changes; when the
brain is diseased, thought is disordered, and when that
ceases to act thought ceases to be manifest. That it is
not to be supposed that any one particle in distinction
from the others has the power of thought, but that it is
the one simple result of the combined action of the whole,
just as music is the result of the combined action of the
fiddle-bow and the fiddle, or as secretion is the result of
the action of the gland. This, I think, is a fair statement
of the doctrine of the materialists, and of the kind of
analogies by which it is supported.
In reply to this I observe, first, that we have evidence
of the existence of thought without a brain or nervous system, or we have no evidence of the existence and intelligence of God, or of any spiritual being. If there be such
beings, doubtless the principle of thought is the same in us
as in them.
But allowing that we have no such evidence, I am inclined to think that the statement is absurd, for it supposes the whole to have properties which do not belong
24*
281
LECTURES ON MORAL SCIENCE.
to the parts separately. In a piece of silver, however
small, we have all the properties we can have in a mass as
large as a mountain.
But music is not the property of the fiddle, or of the
bow, but the result of the combined action of the two.
What, then, is music? What is done in this case? Why,
the fiddle and the bow have motion of a particular kind in
their particles. This is communicated to the atmosphere,
producing vibrations. These vibrations are no more one
than the "gales of Araby the blest," and of Lapland.
They proceed to the ear, and by means of that make a
series of impressions upon the mind which we group under
one name and call music. The only one effect produced
by the fiddle, or the bow, or both together, is motion, and
this is a property or result that belongs to all the parts.
There is no music till the motion reaches the perceiving
mind, which, having an antecedent unity, makes to itself a
unity of that which, without it, had been nothing but a
succession of different motions. Here certainly is no new
property acquired by the aggregation of parts, no unity
like that of thought, nor indeed any unity at all except
that which is derived from the mind itself. A man might
as well speak of the unity of the particles which cause
smell because they produce one odor, as of any unity there
is in music till it reaches the mind.
But does not the brain secrete thought as the liver does
bile? This is a favorite theory with some physiologists.
To this there are three objections: 1st. The liver does
not secrete bile as mere matter. A dead liver will not do
it, and if there were not some one principle of life, different from matter, working through the liver, it would not
do it. 2d. The bile that is secreted is made up of separate particles of matter, and has no unity as thought has.
282
THE MIND AND THE BODY.
3d. Thought is immaterial, and it seems absurd to suppose
that an immaterial result can be secreted by a material
organ.
Than the analogies just given I know of none stronger.
Particles of matter may be so accumulated and arranged
as to produce an effect upon the mind or upon other matter different from that produced before, and may require a
new name, but they thus get no new quality or property
like the power of thought. Here is a single particle. It
has, by the definition of matter, magnitude, figure, mobility,
and if any one shall clhoose to add color I will not now
object. Now, you may add other particles to this in any
way you please, and unless you change the definition of
matter you can have nothing but varieties of color, figure,
motion, and magnitude.
But it is said that the mind and brain increase together,
are mutually affected, and decay together. This is true to
a certain extent, but with such exceptions and limitations
as to destroy the force of the argument. "It is certain,"
says Lord Brougham, "that the strength of the body, its
agility, its patience of fatigue, indeed all its qualities, decline from thirty at the latest, and yet the mind is improving rapidly from thirty to fifty, suffers little or no decline
before sixty, and therefore is better when the body is enfeebled at the age of fifty-eight or fifty-nine than it was in
the acme of the corporeal faculties thirty years before. It
is equally certain that while the body is rapidly decaying
between sixty or sixty-three and seventy, the mind suffers
hardly any loss of strength in the generality of men; that
men continue till seventy-five or seventy-six in the possession of all their mental powers, while few can then boast
of more than the remains of physical strength, and instances are not wanting of persons, who, between eighty
283
LECTURES ON MORAL SCIENCE.
and ninety, or even older, when the body can hardly be
said to live, possess every faculty of the mind unimpaired.
The ordinary course of life, therefore, presents the mind
and the body running courses widely different, and in
great part of the time in opposite direction; and this
affords strong proof both that the mind is independent of
the body, and that its destruction in the period of its
entire vigor is contrary to the analogy of nature." Of the
above statements Lord Brougham is himself a distinguished
example.
No doubt we are intimately connected with certain portions of matter. We are so with our limbs; but cut them
off and there is no loss to the mind. Yet these were portions of matter by means of which we had felt and communicated with the external world. Our connection with
the brain may be more intimate, but there is no reason to
suppose it to be of a different kind. Parts of the brain
may be ulcerated, removed by operations or by accident,
and the man still remain the same. Indeed, there is no
reason to suppose that one piece of matter by means of
which we perceive is any part of ourselves more than
another. By means of glasses we see objects that we
could not without them. The same is true of the eye.
The eye is only an optical instrument which we carry in
our heads instead of our pockets, and we have no more
reason for supposing it a part of ourselves than we have
for supposing a telescope a part of ourselves.
As to the decay of old age and the effects of disease
and injury upon the brain, they are only what might naturally be expected. To adopt the words of an old English
poet, Sir John Davis, -
"For these defects in sense's organs be,
Not in the soul, nor in her working might;
284
THE BODY AN INSTRUMENT.
She cannot lose her perfect power to see,
Though mists and clouds do choke her window light.
" These imperfections, then, we must impute
Not to the agent, but the instrument;
We must not blame Apollo, but his lute,
If false accords from her false strings be sent:
"As a good harper stricken far in years,
Into whose cunning hands the gout doth fall,
All his old crotchets in his brain he bears,
But on his harp plays ill, or not at all."
The most skilful singer may have a cold, the best musician a cracked fiddle, the best eye can see but imperfectly
through furrowed glass. Speaking of this connection and
distinction between the agent and the instrument, Cicero
says: "Suppose a person to have been educated from his
infancy in a cottage where he enjoyed no opportunity of
seeing external objects except through a small chink in
the window-shutter, would he not consider this chink as
essential to his vision, and would it not be difficult to persuade him that his prospect would be enlarged by demolishing the walls of his prison?" You see. the application.
Old age is the gradual closing up of this chink, but death
is the pulling down of the walls of his cottage and letting
in the broad daylight upon him.
From a consideration, then, of the nature of the soul, so
far as we can judge of it from its attributes, we believe it
has an existence independent of the body, and that it is of
such a nature that it can perish only by annihilation, which
we have no reason to suppose ever occurs in regard to the
most inconsiderable of the works of God.
We now pass from the argument from the nature of the
soul as a substance, and its connection with the body, and
285
LECTURES ON MORAL SCIENCE.
proceed to others derived from its faculties and situation
in this life.
The first that I shall mention is the general and wellnigh universal belief of this doctrine. Travellers have said
that they could discover no traces of this belief among
certain tribes. But those who pass through a tribe ignorant of its language and customs are incompetent judges
on such a point, and it has, I believe, been found in every
instance, after fuller investigation, that such a belief did
exist. If it do not, it is only among those who are raised
but one degree above the brutes, and whose faculties consequently have not been developed. Certainly the belief
is so universal that it must be supposed to be the work of
nature. It arises directly and without reflection from our
natural desire of continued existence, and from the expectations excited by the action of the moral nature in hope
and remorse.
Can we, then, reason from the constitution of our nature
to its destiny? -from the expectations which that nature
instinctively excites to our future condition? If not, we
can reason from nothing, for nature is not constituted on
the principle of good faith. It is said, indeed, that the
above principles subserve beneficial purposes in this life,
and that they were given for that purpose alone. But if
so, nature has mingled them in the constitution too largely.
She has so constituted them that man does, in fact, expect
a future life. Besides, if they were given for this life only,
why is hope strongest in the aged and the good, and why,
especially, does remorse increase in the guilty as death approaches? The appetites were given for this life, and they
grow weaker with age. What must be thought of the
h)nesty of a system in which it should turn out that the
286
WANT OF ADAPTATION.
hope of the good man was a lie, and the fear of the bad
one a phantom?
A second argument is, that while there is throughout
nature an exact adaptation of everything else, especially
of every animal in its structure and instincts, to its situation and end, if man is to exist in this life only, we find no
such adaptation either in his intellect or in his affections.
While the brutes have no curiosity, instinct supplying
the place of experience and of investigation, man wishes
to know not only the use of things, but their nature, and
that not alone of things with which he is or can be connected in this life. His curiosity fixes upon bodies the
most remote as if they were his own proper province
which he was one day to investigate and understand. He
sees the mountains and valleys of the moon; he follows
the track of the comet; he wonders at the rings of Saturn;
he explores the nebulae, and inquires after the "architecture of the heavens." He knows just enough of these
bodies to raise his curiosity, but not enough to satisfy it.
His intellect is to the distant universe just what the eye
of the child is to this world when it first opens upon it. It
sees a little, it is adapted to see more; shall it be quenched
forever even before it has learned to see? And not only
does the intellect seek to know the physical universe,
it also inquires after God. It says, "Where is God my
Maker?" It is capable of knowing that God as seen in
dim reflection from his works. And will God, having revealed himself thus dimly, withdraw? The twilight of
this highest of all knowledge having dawned on the soul,
shall the sun go back?
Nor is the want of adaptation in the affections less than
in the intellect. Even more than curiosity do the yearn
287
LECTURES ON MORAL SCIENCE.
ings of affection seem like those mute promises of nature
which we observe in the animal world, and which are
there always fulfilled. Has the chrysalis wings that are
folded within? There is an atmosphere prepared without in which it can fly. Does the bee perceive a fragrance
on the air? There is a flower on which it can alight.
And shall man be as the bee that should perceive the
fragrance but not find the flower? Shall bereaved love
cherish hope till death only to be disappointed? The
only interpretation of affection yearning for something
higher than this world can give, and the only solace of
that which is bereaved, is to be found in the doctrine of a
future life.
A third argument is, that while individuals of every
other species attain all the perfection of which their nature admits, there is evidently a foundation laid in the
nature of man for an indefinite progression.
A tree rises from its seed, it increases for many years, it
is beautiful to the eye, it yields fruit, it furnishes shade.
If it were to remain forever it could do nothing worthier
or better. It has attained perfection as a tree. So an
animal reaches in a short time the limit of its powers.
Destitute of reason, of a moral nature, of the power of
forming general ideas and following general rules, it goes
on in the fulfilment of its destiny guided by instinct and
by particulars, and could not, without a different kind of
powers, make any essential progress. It is perfect as an
animal. The structure, though humble, is complete. It
has its capitals and its dome. But no one can say that
man, considered as a rational and moral being, reaches
here the perfection for which a foundation is laid in the
nature of his powers. The philosopher who has traversed
288
ARGUMENT FROM ANALOGY.
the circuit of human knowledge, and has pitched his tent
upon its outposts, not only does not approach the limits of
knowledge, but, what is important to our present argument, he does not find his powers burdened or embarrassed by the knowledge already acquired. On the contrary, every advance which he makes gives, and from the
nature of the powers must give, new light and strength to
make further advances, and when old age comes he only
feels himself more "like a child gathering pebbles on the
shore of the great ocean of truth." So, also, and more so,
is it with the good man making progress in goodness. His
path is like the shining light. Shall it shine more and
more unto the perfect day, or shall it go out in darkness?
Here, then, the foundation is laid. Shall the superstructure go up? The ocean is before man, shall he embark
upon it? Or shall he, who, as Shakspeare says, is "'so
noble in reason, so infinite in faculties, in form and moving so express and admirable, in action so like an angel, in
apprehension so like a God," - shall he be left the only
fragmentary being, as if God had completed everything
else and had failed in his grandest undertaking; as if he
had indeed made him not only the "glory," but the jest
and riddle of the world?
The force of this argument from the nature of the
human powers as progressive, is greatly heightened when
it is considered in connection with a fourth which I now
adduce, and which is from analogy.
The effect of this argument upon our minds will depend
much upon the care with which we have studied the
works of God, and our consequent conviction of the connection, and uniformity, and consistency of those works.
It was once supposed that different parts of the earth were
25
289
LECTURES ON MORAL SCIENCE.
governed by different powers and laws. The heavens had
their Jupiter, and the ocean its Neptune. Ignorant nations think so still. But we now know that it is all one
system of mutually dependent elements, and subject to
one code of physical laws. And not only so, it is now
settled that the physical universe is one great system in
which every part is related to every other part. Every
particle of matter, or, as Paley says, "the chicken on its
roost," is related to the sun and to other systems; and
while this is so, we say that it is contrary to all analogy
that this great intellectual and moral system of man should
have no relations beyond itself. The inference undoubtedly is that as matter is here subordinated to mind, it is so
elsewhere, and that there is a vast intellectual and moral
system, the parts of which have relations that are to be
unfolded in future time. I believe that we shall one day
know the history of other worlds and other orders of beings,
and that they will know ours. Certain it is, however,
that we are in the midst of a system; it is one of gradation and mutual dependence from the infinitely minute up
to man, each that is below being united to that which is
above, till we come to man, who is the topmost of the
series. Now, does the series stop with him? Is it probable, when we see, as we do see, the hooks and grapplingirons fastened into the centre of his being, by which he
is to lay hold on somewhat above him, that there will be
nothing found on which he may fix? Our superior knowledge of nature in its connections gives us a great advantage
over the ancients in this analogical argument. It enables
us to see more completely what is wanting that man may
be wrought in and form, without discrepancy, a part in the
one great system; and I would sooner believe in such monsters as centaurs and satyrs and hippogriffs, than in such
290
UNREDRESSED WRONGS.
a monster as man would be if cut off from anything higher
than himself and beyond the present life.
" Unless above himself he can
Erect himself, how mean a thing is man."
But the argument which has probably been most effective with men in general is that indicated in the soliloquy
of Cato:
"If there's a power above us, And that there is, all nature cries aloud
Through all her works,- He must delight in virtue;
And that which he delights in must be happy; -
But when? or where? This world was made for Cesar."
"When," says Jeremy Taylor, "virtue made men poor,
and the free speaking of brave truths made men lose their
lives, or at least all the means and conditions of enjoying
them, it was but time to look about for another state of
things where justice should rule and virtue find her own
portion."
No thoughtful manycan contemplate the unredressed
wrongs of this world without perplexity. He cannot but
ask if there be not a supreme power somewhere who will
regard the appeal of the martyr which he writes in his own
blood,- if there be not a supreme power somewhere who
will disown the atrocities perpetrated in his name on the
field of battle and in the dungeons of the inquisition.
For the difficulties thus presented there is no solution but
in a future state.
What has just been said goes on the supposition that
the distribution of good and evil in this world is promiscuous; but if it should appear that it is not wholly so, but
that there is an undertone in the natural government of God
that chimes in with the voice of conscience, the argument
291
LECTURES ON MORAL SCIENCE.
would be much strengthened. Are there discoverable
amidst this apparent confusion the beginnings of a right eous administration, which carry with them a promise of
their own completion?
This is a subject of wide compass, and for its full illus tration I must refer you to a chapter upon it in ButleF's
Analogy. Suffice it to say, that from the serenity of virtue
on the one hand, and the uneasiness of vice on the other;
from the different treatment which virtue and vice, as
such, necessarily receive in civil society, the constitution
of which is natural; from the forebodings of conscience;
from the natural tendencies of virtue and vice,- the one
tending to order and strength, the other, though it may
have at times an accidental ascendency, to disorder and
confusion,-from all these He who is supreme in nature
has sufficiently indicated to which side he belongs. His
object seems to be to manifest himself just so far as to
give room for moral election, - to give such indications
of his will as may suffice for the sincere, the humble, and
the diligent, but not such as shall be obtrusive, and withhold the reckless, through fear, from pursuing their own
course.
But whatever the object may be, certain it is that the
moral feelings of man are not the only ground of argument on this subject. There is a righteous administration
already begun here, and on the scroll of Providence, as it
is unrolled in its grand and solemn movements, there are
written characters, which vice, if it were not infatuated,
would read and tremble. If, therefore, there be no future
state in which these silent prophecies may be accomplished, then is there falsehood inscribed not alone on the
intellectual and rational powers, not alone on the mere
natural government of God, but also upon his moral gov
292
NATURAL CHIANGES.
ernment so far as it can be discovered, and in the very
sanctuary of the moral nature of man.
If, after these arguments, there should still remain some
vague impression that in the shock of so great a change as
that of death the principle of thought should not survive,
there are analogies of nature which may bring us some
relief. If all the philosophers on earth had been shown an
egg for the first time, and been asked what it would become,
they could as little have thought it possible that it should
be such a creature as a swan or a peacock, as the greatest
skeptic now thinks it possible for man to survive. Or, to
take a case sometimes thoughlt to be more in point, what
can be a greater change than the chrysalis undergoes in
its manner of life, when it passes from its dormant condition to that of a beautiful butterfly, seeming, as Bryant
says, "a living blossom of the air"? So striking, indeed,
is this analogy that the Greeks gave the soul the same
name as the butterfly, from the expectation that it would
undergo a similar change.
The strongest case, however, is that put by Butler. It
respects the change in man from the mode of his existence
before birth to that which he at present enjoys. He is
still the same being, but his mode of existence was so
different that had he been endowed with the powers of
reason he would have been much less able to form any
conception of his present mode of being than he now can
of a future state. Hie might have perceived some indications in his structure, as in the eyes and the lungs, of a
preparation for a state then future, as we now do for one
still future, but the necessary change would have been
quite as mysterious as that which must pass upon us at
death.
It is to be remarked, also, in thinking of this change, in
25*
293
LECTURES ON MORAL SCIENCE.
what part of our nature it occurs. We have two modes
of being, that of sensation and that of reflection, which
seem in a great degree independent of each other. Reflection, having once commenced, is independent of sensation, and is most active and intense when sensation is
weakest. If we wish to reflect we shut out sensation.
But it is upon the sensitive life that this shock of death
seems to spend itself. The power of reflection often continues in full vigor up to the last moment. Since, then,
the power of reflection is so independent of the sensitive
life, and of the organs of sensation, it seems rational to
conclude that it may hereafter maintain a separate existence.
Such are some of the arguments drawn from nature
which I would urge in favor of the probability of a future
life. To me they seem to have no little weight. But if
they were less forcible than they are, so that their opponents could bring against them those of equal or greater
force, I could never understand, unless something different
from mere argument is concerned, the triumph which some
men appear to feel when they suppose themselves to have
quenched the hope of man, and to have levelled themselves with the clod. Surely, if a man were to think himself obliged, as the result of a candid investigation, to
believe that to be true which "nature never told," we
should expect, instead of exulting, that he would
"Read, nor loudly nor elate,
The doom that bars him from a better fate,
But, sad as angels for the good man's sin,
Weep to record, and blush to give it in."
These few arguments from nature for a future life I
offer, not as affording absolute proof, but a presumption so
strong that a prudent man might act upon it even if he
294
CONCLUSION.
had no other light; a presumption stronger than that upon
which we often act, and upon which it would be madness
not to act in the ordinary concerns of life. I wishi, also,
to have you see that skeptics may be met on their own
ground, and that no impression may be left upon your
minds which shall prevent you from receiving in its full
force the evidence for that revelation by which alone, in
all its clearness, "life and immortality are brought to
light."
295
SUMMARY.
HAVING completed the several lectures, it remains to
give a summary of the course of thought passed over.
That course has been one, and in itself entirely simple.
The three questions proposed concerning duty-1st.
What ought to be done? 2d. Why ought it to be done?
3d. How ought it to be done? - we attempted to answer
by a consideration of ends. We saw that all rational arrangement, construction, and action, must have reference
to an end, and can be comprehended only in the light of
that end; and that all rules and laws have their significance and value in the same way.
We assumed that from a study of the structure of man,
physical and mental, some knowledge may be gained, not
only of his separate organs and faculties, and of their use,
but also of the end of man himself. If man cannot know
his own end there can be no philosophy of man- no comprehensive or satisfactory knowledge of him. Whether he
could know this as he now is, without revelation, may be
doubted. There is no philosophy in a ruin; and where
the Bible has not been it does not appear that men have
retained a knowledge of their end. But however this may
be, a knowledge of the end must greatly aid us in tracing
the arrangements and correlations for the attainment of
296
SUMMARY.
that end, and so of comprehending the whole system as
one of means and ends.
Ends were distinguished as subordinate, Ultimate, and
supreme.
As the conception of an end involves that of some good,
we considered the nature and sources of good. This we
found to result from activity, and that the highest good
would be from the activity of the highest powers in a
right relation to their highest object. We discriminated
the different kinds of good as it comes from the susceptibilities and the powers, finding fiom one what is distinctively pleasure, fiom the other happiness and blessedness.
We then sought to classify the powers, and consequently
the good derived from their action, as higher and lower.
In doing this we found a common law of gradation, and so
of activity for forces and faculties - — for those forces by
which the universe is governed, and for those faculties by
which man ought to be governed..
Commencing with the lowest and most general force
known to us, we passed up till we came to vegetable or
organic life, where a great transition is made, and which
subordinates to itself all lower forces. We then came to
that sensitiveness and intelligence in the service of which
life works; and then to those rational and moral powers
in which is personality, and by which we are made in the
image of God. At every step from the lowest sensitiveness, while we found, as in that, an end in itself, we also
found a beautiful subordination to that which is higher,
and in that subordination we found the law of limitation
for the activity of every lower power and faculty. We
saw how perfectly God regards this law in that part of the
chain where our wills do not intervene, and how perfect is
the model he sets before -us for the regulation of' our own
9
29T
SUMMARY.
lives. We saw that when we reached the highest form of
activity and of good the law of limitation ceased, and became that of the highest capacity of the faculties in a
form of activity, and so of blessedness, like that of God
himself. Our conception of him is that he is perfectly
blessed in a holy activity. Being made in the image of
God, our whole duty and end, as might have been supposed, is to be like him; and if we are like him in his
activity we must be in his blessedness.
In thus passing upward from a broader basis, retaining
all that is below, and adding something for every new and
narrower platform, till we reach man at the summit of the
pyramid, we find for the universe so far as we know it, the
principle of unity. This is in the fact that each lower
force is always the condition of the higher. This would
give us a universe; but it is the fact that each lower force
is precisely such in degree as to be the most favorable condition for that which is higher that gives us an orderly
universe. This fact and relation we find everywhere in
nature, -in all the systems of which the body is composed, from the digestive upwards, and in all the powers
and faculties of the mind; and everywhere we find the
proportion of force accurately observed till we come to
the intervention of finite will. We thus find provision
for every inferior form of good. We omit nothing; we
undervalue nothing. We find provision for the harmonious operation and symmetrical growth of every propensity, appetite, and power, whether of body or of mind, and
especially full provision for those powers by which man is
connected with what is infinite and eternal.
Having thus obtained a knowledge of good in its
sources and gradations, we proceeded to a classification
of those activities and faculties in which good originates.
298
I
0
SUMMARY.
And here we considered the forms of mental activity,
first as spontaneous, and second as voluntary. In the first
we found a spontaneous or automatic life which is conditional for a voluntary life, into which the voluntary or personal life is put, as into a garden, to dress and to keep it,
and which, without the personal life, would go on always as
the mind does in dreams, and be a thinking thing. Perfection here would be in the coincidence of the two without effort.
We then proceeded to classify the faculties as they are
related to ends.
Here the first class is of those which are instrumental
for the attainment of ends beyond themselves. Under
this are, first, those which indicate ends; and, second, those
in the light of which we pursue ends.
The second class are those powers in whose activity we
find ends beyond which there are no others, and which are
our moral nature.
Under the first class we considered separately, and at
some length, the Instincts, the Appetites, the Desires, and
the Natural Affections. We considered the Intellect, also,
as far as that is subject to the will. All these we regarded
as having no moral quality in themselves,- as neither
good nor evil, except as they are controlled. To any particulars of these discussions which occupied us during four
lectures we need not now recur.
We next passed upwards to those powers that are directive, that are our moral nature, and in whose activity are
ends beyond which there are no others. These we found
to consist of the moral reason, having inseparably connected with it moral affections and conscience; and of
free will. In the union, or rather synthesis of these, we
found a person, and so reached the highest known and
22*
299
SUMMARY.
possible form of being. Not that the person is composed
of these as elements, but that this person is as a simple
form of being, and that these are forms of its manifestation without which personality could not be conceived of.
Here we find a being moral and responsible.
In the activity of such a being, naturally knowing his
own end, and necessarily affirming obligation to choose it,
we have the intuitional side of a true moral system; and in
the activity of the discursive and practical powers in coincidence with this we have its inductive side. We thus
harmonize intuitional and teleological systems. In this
connection, also, we have the characteristic of complete
virtue or holiness as manifesting itself in two directions;
we have the point of moral responsibility, and the genesis
of our chief moral ideas. Here, too, we considered the
moral nature in its double function, as both originating
moral acts, and judging of them; and here we sought for
the proper sphere of conscience, and pointed out the ambiguity of the term.
Of a person thus endowed with reason, moral affections,
conscience, and free will, the highest form of activity is
rational love; and hence, according to the philosophical
formula for the highest good, we found it here. At this
point, therefore, we identified the teachings of the human
constitution, as drawn from a consideration of ends, with
the summary of the revealed law of God as given by our
Saviour.
We next investigated the relation between holiness or
virtue, and happiness. In doing this we distinguished between moral good, as the natural and necessary result of
moral goodness, and natural good; and also considered the
good there is from the approbation of goodness. Moral
good and that from approbation were shown to be infal
300
SUMMARY.
libly connected with moral goodness. Natural good is not
necessarily thus connected, but there is a tendency towards
it. There is between them no contrariety or opposition,
or "antinomy," and they ought to be connected by will
in the way of reward. That they are not thus connected
in the present state, is an evidence of disorder, and an
indication of a state yet future.
In connection with this we affirmed the duty of each
one to secure his own good through moral goodness, and
found that this was not only compatible with the good of
the whole, but necessary to it,- thus bringing into harmony a rational self-love and benevolence.
Regarding not only the quantity, but also the quality of
enjoyment, we saw that the good and end for man was
not to be found either in holiness by itself, or in happiness
by itself, but in holy happiness, or blessedlness. That these
are thus necessarily united, no doubt God intended we
should know; also that we should seek them as thus
united; and our idea of perfection is the highest possible
union of these, together with all natural good following in
their train.
In determining, next, more specifically, the sphere of
moral science, we took our point of observation at the
performance of an outward act, and going backwards to
its source, we found an immediate recognition of the
moral quality of the act as good or evil; while, in going
forwards and outwards to its consequences, we found the
ideas of utility, and, in one sense, of right and wrong. In
the one case we were wholly concerned with the person
and the motive; in the other, with the outward act and
its results. Separated from its origin in a person, and its
motive, an act can have no moral quality; but it may
be outwardly conformed to law, and have consequences
26
301
SUMMARY.
beneficial or injurious, and be, in ordinary language, right
or wrong; and an attempt was made to show the confision that has arisen at this point, and the need of greater
precision both of ideas and of terms.
We also considered the province of conscience, its infallibility, the two spheres inll which it acts, and its relation
to other active principles; and we inquired whether, in
order to be virtuous, an act must be done from a sense of
duty.
Leaving personality and motives, we next went outward
to the consideration of those fixed relations established by
God, and which indicate his will. Here we saw that virtue and rectitude are so far coincident that where virtue
exists there can fail to be rectitude only from mistake;
and also the difference between those calculable consequences from acting in violation of fixed relations or in
accordance with them, and those incalculable and illimitable consequences that may flow from guilt or its reverse.
We sought the character of a true expediency, and the
difference between prudence and virtue. We even ventured to speak of the nature of God, and so far to call in
question the common view as to suggest whether it be not
his nature to be wholly supernatural; and whether there
can be anything more ultimate for the conscience than his
character as the standard of moral excellence, and his will
as the expression of that character.
At the opening of our discussions it was said that besides pursuing an end as rationally comprehending it, we
may also do so from Instinct and from Faith; and we next
showed that between the action of these and of reason
there might so be a coincidence that a man may be rational in acting both fiom instinct and firom faith. Reason
and faith being thus reconciled; reason being at the basis
302
SUMMARY.
of moral philosophy, and faith being the distinctive principle of religion, just as it is in the relation between parent
and child, it was easy to see what must be the points of
coincidence and mutual support between moral philosophy
and religion, whether natural or revealed- whether a system of pure revealed law, or of forgiveness and restoration
after law had been broken.
We next had before us the subject of rights as connected with our previous speculations. We showed their
origin in the will of God -uttered through the several
active principles of our nature- that man should attain
his end. We ascertained their gradations as growing out
of previous classifications. We drew the distinction between alienable and inalienable rights, and also between
those over persons and over things. We showed the
foundation and limits of the rights of parents and of governments. We spoke of liberty in its various kinds as
related to rights; also of the rights of different classes of
the community; and closed by a reference to the duty of
all in a government like ours to secure the rights of all.
In the closing lecture we have passed from the relations
of time, and considered the great question of a future life,
thus giving to morality weightier sanctions, and a loftier
perspective. The details of the argument we need not
reproduce.
We have thus, my friends, in accordance with that
ancient precept, "know thyself," which is said to have
descended firom heaven, examined the human constitution
in its relation to ends. In doing this it has been my wish
to avoid technical terms, and to appeal directly to the consciousness of my hearers. That appeal has been met by
an attention that has been all I could desire. Upon such
a course probably no independent thinker could enter
803
SUMMARY.
without discovering new relations both between the faculties themselves, and between them and the ends for
which they were intended. Hiow far such relations have
now been presented, or the point been reached towards
which the great lines of thought converge, you will judge.
That these views will be accepted by all, I do not expect.
That they will not be without their value in advancing
the science, I cannot but hope. As was said in the first
lecture, that advance must be slow; but we are not to be
discouraged. The moral sphere is more intimate to us
than any other; it is the highest of all; it is there that we
find our true selves; and it cannot be that we should be
capable of tracing the harmony of suns and of planets,
and be forever incapable of apprehending those higher
harmonies which we have now attempted to trace, between man and nature, between man and himself, between
man and his fellows, and between man and God.
THE END.