Panentheism.  Revisionism.  Anarchocapitalism.




Essays by Me

Essays by Others

From The Review of Metaphysics, Vol. 36 (1982), pp. 23-49, reprinted in his Individuals and Their Rights (Open Court, 1989). I thank Professor Machan for providing me with this slightly revised version. 


Epistemology and Moral Knowledge

Tibor R. Machan



Arguments abound against the view that it is possible to know what is morally right or wrong, good or evil, in personal, social, political, and other realms. Such arguments, as well as those that defend the possibility of moral knowledge, have had varying degrees of success throughout the history of philosophy.1 In recent times, the skeptical stance has fared quite well, whereas in the era of Plato and Aristotle, for example, cognitivism—in the form of essentialism or naturalism2—held the upper hand.

The Platonic-Aristotelian approach in meta-ethics seems once again to be appealing to a significant number of working philosophers. But while a resurgence of naturalism has emerged on several philosophical fronts, not much has been done to synthesize it, to assemble its various strands. This article will present, in outline, a position that both gives support to and makes use of a certain rendition of the argument that we can identify right and wrong, etc., by reference to the nature of things, that is, by considering what something is essentially or by definition. Not all the various argumentative strains involved will be fully developed, but the direction of each argument that is relevant will be clearly enough indicated. Thus the prospects of the various strains, as well as of the entire position, will be available for scrutiny. This sort of synthesis seems to have its importance, as much of traditional philosophical thought suggests.

A prominent view of what a theory of knowledge must accomplish will first be challenged since this view has made much difference in the present endeavor, viz., basing morality on the nature of things. Next it will be suggested why the requirements placed on a theory of knowledge have made it appear that we can know what is the case in non-normative domains but not when it comes to norms—or principles of right and wrong—concerning human conduct, institutions, laws, etc. Here the source of the famous “is/ought” gap will be indicated. A contextualist account of human knowledge will be proposed and it will be shown why it is quite possibly correct, suggesting how this approach to understanding knowledge could work toward eliminating the dichotomy between “is” and “ought.” If what is maintained in this paper is indeed correct, the possibility of moral knowledge will (again) be seen as a reasonably good prospect. And finally, such a prospect will be indicated at the conclusion of this paper.


Analyses of human knowledge have to a considerable extent been under the influence of the belief that stating what knowledge is will, if true, amount to a necessary truth—that is, a formal proposition that is timelessly incontrovertible. Let me spell this out a bit.

Suppose someone defines the concept “knowledge” as “P.” Suppose, also, that a complicated science fiction story can be imagined—or a conjurer’s trick, or a very clever conspiracy—in the light of which it appears that someone has attained a clear grasp of some state of affairs or situation but in so doing has not satisfied the criteria specified in “P.” Our ordinary notion that having grasped the state of affairs—having come to be fully confident of the relevant state of affairs— is exactly what amounts to knowing that state of affairs now conflicts with the definition of knowledge proposed. Accordingly, the definition will be dismissed as a bad one because the reigning model of what knowledge must be requires that when the concept is defined correctly as “P,” the relationship between the concept and “P” can only amount to a mutual entailment.

That this is the reigning model is surely not controversial, although some challenges can be found.3 It will be useful to have a clear cut example of the use of the model in the effort to provide an analysis of the concept of knowledge, lest a mere general statement be construed as perhaps a slight misrepresentation. In his well-known discussion of the topic, Keith Lehrer tells us that he is concerned

with precisely the question of what conditions are necessary and sufficient for a man to have knowledge, or, more precisely, to know that p or that S is true, where “p” is a variable that would be replaced by a declarative sentence and “S” by the name of a sentence.4

Lehrer goes on to say that

an analysis of knowledge may be given in an equivalence of one of
the following two forms:

S knows that p if and only if.


S knows that Q is true if and only if.5

Furthermore, this approach, taken by many others besides Lehrer, is also regarded to be free of any preconceptions or bias. Lehrer states, for example, that:

In our investigation we shall begin without making any assumptions concerning the postulates of our theory of knowledge and thus consider any logically possible case as a potential counterexample.6

In other words, Lehrer seems to believe that his program for presenting “an analysis of knowledge” places no restrictions on what the outcome might turn out to be. In fact, it is clear that Lehrer’s specifications given above are infused with certain assumptions, including the most crucial one concerning what a theory of knowledge must produce, namely, a proposition with a very strictly specified form.7 When Lehrer states that “an analysis of knowledge may be given” in the manner he states, he is restricting any purported conception or theory of knowledge to producing necessarily true statements, ones expressed via the timeless conditional of “if and only if,” a logical connective that can only be satisfied if only purely formal relations of a certain sort are satisfied.

Is it reasonable to require that knowledge be analyzed along lines suggested by Lehrer? Is the built-in timelessness demanded of any purported analysis, definition, or statement of the nature or essence of knowledge something indispensable, as Lehrer suggests by his statement of how an analysis “may be given”? Must we be asking “the question of what conditions are necessary and sufficient for a man to have knowledge” in the formalist sense Lehrer prefers, or does asking the question along such lines already prejudge for us the sort of outcome that will be acceptable? And while there is little controversy about the meaning of the phrase “logically possible,” there is considerable debate about what is presupposed in an accurate account of that meaning.8

While a good deal of epistemological work follows the approach Lehrer uses, some changes are evident. And certain philosophical positions that rely on the results of work in epistemology need not be evaluated on the basis of whether the arguments in support of them conform to the requirements embodied in the approach Lehrer and others take to the analysis of knowledge. That is to say, if one were to claim that it is possible to know some proposition even if knowing it could not be analyzed along the lines Lehrer prefers, there is work in epistemology that could vindicate the claim.

For example, already in J. L. Austin’s “Other Minds”9 a different characterization of knowledge emerged. In terms of his account, it could be said, correctly, that someone knows something while the possibility of some wild counterexample could also be accepted.1° It seems that Austin’s idea of what is required in an account of knowing differs considerably from what, e.g., Lehrer assumes such an account must achieve. Suppose we claim to know that something is a goldfinch and suppose, also, that goldfinches are not talking birds. It does not follow, from Austin’s conception of knowledge, that the possibility of some such bird quoting poems must lead us to withdraw our claim to knowledge. If some goldfinch actually did produce poetic utterances, our concept of gold- finches would have to be revised but all our previous knowledge claims would not be invalidated. For those like Lehrer, however, a definition of knowledge carries with it a claim to logical necessity (and certainty) which then—because they are such strong requirements—admit highly implausible defeating possibilities. Austin’s approach, however, could accommodate a number of areas of conceptual change, leading to a position in line with which we need not accept the view, e.g., that science is a history of hopeless errors because it does not produce timeless truth and knowledge. Austin’s idea is more in accord with Gilbert Harman’s advice that “We must take care not to adopt a very skeptical attitude nor become too lenient about what is to count as knowledge.”11



 In providing an alternative to the formalist approach Lehrer takes in trying to identify what knowledge is, let us recall a very instructive passage from an essay by Barry Stroud in which he set out to explain Wittgenstein’s idea of logical necessity. Stroud’s interpretation of Wittgenstein can set the stage for a non-formalist account of human knowledge. Stroud tells us that by Wittgenstein’s explanation, logical necessity

is not like rails that stretch to infinity and compel us always to go in one and only one way; but neither is it the case that we are not compelled at all. Rather, there are the rails we have already traveled, and we can extend them beyond the present point only by depending on those that already exist. In order for the rails to be navigable they must be extended in smooth and natural ways; how they are to be continued is to that extent determined by the route of those rails which are already there.12

It is instructive to note that Stroud adds to this characterization a point about why he has addressed this issue. He says that he has “been primarily concerned to explain the sense in which we are ‘responsible’ for the ways in which the rails are extended, without destroying anything that could properly be called their objectivity.”13 Stroud indicates by this remark that he disapproves of the equivocation between formalism and objectivism regarding what amounts to knowing the nature of something like knowledge. To put it another way, objectivism does not require Platonism concerning the nature of what is known.

To develop some of the points suggested thus far, a different approach to answering the question “What is knowledge?” will now be provided from that which Lehrer and other formalists have been urging upon us. It bears mentioning at the outset that what follows from this approach applies not only to answering the question about the nature of knowledge but also to answering questions concerning the nature of anything else. In other words, this paper aims to provide the groundwork for a theory of (the character of) definitions. The difficulty here consists, in part, in distinguishing between the general question, “What is (the nature of) X?” and the more specific question of “What is it to know (the nature of) X?” Even more specifically, this paper is also concerned with shedding some light on the proper approach to the question of “What it is to know (the nature of) knowledge?” Although the answers to these questions overlap, they are not identical, and it will be clear shortly why this is so.

Knowledge, to begin with, is just as much part of the natural world as anything else we are familiar with—e.g., chairs, trees, sorrow, explosions, months, melodies, and so forth. Knowledge takes place on this earth, in our universe and is found where we find all those other aspects of reality we want to understand and define correctly. As with the rest, we could benefit considerably from arriving at a definition of knowledge, by answering the general question “What is (the nature of) knowledge?”—that is, by placing knowledge into a coherent and useful scheme of categories. 14 Putting it this way by no means renders the conception of a definition of something either nominal or conventional, linking our definitions less to nature than to our interests. If firefighters are using some method by which to extinguish a blaze, it is true enough that what they are doing relates to their or society’s concerns. But it does not follow that they therefore can go about this task in any which way we might imagine. The fire itself, as something definite, imposes standards on the process of fighting it. Similarly, while it is true that human beings are concerned to understand reality, to make distinctions and to integrate things in a coherent fashion, reality itself presents for them limits on their options and methods, as regards both what there is to understand and how they can go about doing this task.

Since we encounter knowledge in nature, it is reasonable to expect that knowledge would share some of the features or characteristics we find true of other aspects of the natural world. For example, it is temporal and spacial.15 Of course, which aspects of knowledge are temporal, and which spacial, are complicated issues. All that is suggested here is that knowledge is to be construed, in its most general respect, as a natural phenomenon. As a consequence, it is possible for knowledge to undergo some changes as time passes and all the activities and things surrounding knowledge exert influence on it. So understanding knowledge would suggest right off that a purely formalistic approach to understanding it would be off the mark.

When seeking a definition of knowledge, it would be wrong to preclude the possibility that knowledge can change, although the conditions for such change could be quite rigorously circumscribed. But the possibility of having to change a definition developed by our best lights today would have to be admitted. This, in turn, leaves it open that such a definition would not manage to withstand the test of counterexamples based on what is only imaginable or possibly possible.16 In terms of the approach suggested here, defining knowledge—giving an analysis or developing a theory of it— would have to be regarded as a task similar to those performed by botanists or geologists or astronomers, all of whom aim at formulating definitions without assuming the impossible task of making those definitions timelessly correct.

In contrast, then, to Lehrer’s approach, here definitions are not modeled on necessary truths. Yet this framework does not preclude the possibility that some definitions will be necessary truths. Inquiring into the nature of something involves leaving open the question whether what will be discovered or formulated will or will not have the character of necessary truth—a point that would seem to be objectionable only if it is already taken to be necessary that definitions are necessary truths (or, alternatively, cannot be necessary truths). Some definitions would, accordingly, be open- ended—e.g., the statement of the central characteristics of some little-known class of items in astrophysics, say, black holes. Here further inquiry will probably require considerable readjustment, yet even the most thorough study will not yield a definition that can foreclose all further revisions. In metaphysics, on the other hand, it is mostly likely that definitions of key concepts will have to be necessary truths as well, because if metaphysical propositions can be true, they would hold for the past, present, future, and the possible—that is, for everything at all times.17

A science—and, at the outset, every budding scientist—begins with encountering nature in the rough and ready, so to speak.18 Only later do detailed studies get under way. Similarly, epistemologists first encounter knowledge in the rough and ready and only later endeavor to embark upon formulating a definition of their subject. Epistemologists, as other students of nature, must leave open the possibility of being warranted to formulate a definition of knowledge which conforms to standards of adequacy which differ from standards applicable in other fields, including in other branches of philosophy, e.g., metaphysics. (Yet, it seems it has been their initial commitment to finding metaphysical truths, on which to rest others in the firmest possible fashion, that has led them to embrace the standards applicable in metaphysics in their search for truth about the nature of knowledge.)19

The suitable approach for epistemologists would seem to be to look around for cases of human beings trying to become better and better aware of the world, more and more surefooted in what they claim to be aware of, and so forth. They should study human beings who embark upon comparisons of this with that, upon differentiation and integration of what they encounter in nature. From this study they should then develop a theory of knowledge—of what it is to know—in general, and what differences there might be found in the knowledge of differing aspects of reality. Along the way, the epistemologist would very likely find that knowledge can be obtained of not only different kinds and types of things, but also of things in different respects—e.g., in particular and in general. This would probably lead the epistemologist to the development of a theory of definitions. As the epistemologist attains, at any given time, the most coherent and complete conception of what knowledge is and states the results in an organized proposition, what is obtained is a definition of knowledge.

Now clearly there are some self-reflexive elements in this process, since as the epistemologist is learning about knowledge, including about knowledge of definitions, one of the results that will emerge is knowledge of the definition of knowledge. This is a complication and requires extreme care, admittedly. And there is not going to be any guarantee against error. But the last thing that an epistemologist should demand is that the definition of knowledge produced must withstand a test that is properly aimed at scrutinizing timelessly necessary truths. Yet it is precisely because of this kind of test that definitions of knowledge tend to succumb over and over again, especially in such difficult areas of knowledge as ethics and politics. But we will turn to that later.

Let me note here that what has been suggested above may have two very important results. Just as the conception of knowledge is illegitimately modeled on necessary truth, so the conception of definitions in general must be freed from that prerequisite. Instead, the contextually correct definition (given time and present information) is what can and should (only) be required. This suggests that the most rationally warranted identification of a place in a scheme of categories that has been developed (by means of our sorting, listing, grouping—i.e., study—as we differentiate and integrate the available materials in some range of awareness) qualifies as the definition which should be formulated. The definition of knowledge itself would best be conceived of as the currently most rationally warranted end result of the epistemologist’s study of a given range of phenomena. But this characterization of definitions would also be part of the epistemologist’s stock in trade. That is to say, one of the results of a sound theory of knowledge would be that when someone knows the nature of something—i.e., has formulated a correct definition of this something, say, the concept “knowledge” or “human being” or “justice”—this would enable one to produce a statement which correctly places some range of items, events, activities, institutions, or the like in the most up to date, well developed scheme of categories human beings have managed to develop.

Before some additional material to support the present approach is provided, some possible objections need to be addressed. First, it may be argued that to abandon the possibility of timelessly true definitions is to abandon also the possibility of correct or true definitions, since without the prospect of such timelessness, the generality of our definitions (and the understanding supposedly facilitated by them) will vanish. After all, one point of defining a concept, for example, and “knowledge,” is to have some initial understanding of any instance of, say, someone knowing something—i.e., so as to be able to distinguish a mere claim to knowledge from a genuine instance. Yet without the timeless quality, what would justify our retaining any definition from one instance or case to the next, when all these take place at a different time from the next and so forth?

Another challenge that is worth meeting right off is that which would identify the present position with out and out nominalism. If definitions are not timeless and, indeed, are places in schemes of categories, what guarantee do we have that they state the nature of things? What relationship do such definitions—such purported statements of the nature of something—have to reality, to the actual beings of things, events, etc.? Is it not more likely that conceiving of definitions or the nature of things along such lines drives us toward the view that what definitions are comes to no more than labels or names which comprise, at best, a consistent system with no established tie to reality at all?

More problems could be discussed, but these are the most crucial ones to handle. First, then, let me note that the requirement of timelessness, although one with a long history, is unreasonable even though suggestive. That is to say, although strict timelessness is an unreasonable standard to set for the adequacy of correct definitions, stability through different spans of time is definitely not unreasonable. As Stroud notes, there must be something workably steady, smooth, and natural in the way in which the definition of a concept—or the necessary and sufficient attributes of something—evolve through time. Different aspects of nature, of course, possess different degrees of stability, durability, integrity, and so forth, and a correct definition of such different aspects of nature will need to take careful account of that fact. A definition of the concept “human being” that makes it impossible to integrate our knowledge of human life today with the lives of the men and women of ancient China or Egypt will falter on grounds that the evidence we have of these civilizations will be rendered incoherent without a more durable and comprehensive conception of human nature. What is unreasonable, however, is to require that a correct definition of human beings should have to accommodate every logically possible fantasy that might be imagined in connection with something roughly akin to human life. The sort of science fiction cases often used to test ethical and political theories—whereby some definition of human justice is put to the test by reference to the possible or possibly possible behavior of “human” beings that have no personal identity or any interest in their children or personal safety—testify to the adherence of some philosophers to a conception of timelessly true definitions which sees the world not so much as a stable place but as a static one. In short, then, contextually true definitions will need, still, to do justice to the actual stability of the phenomena of the natural world and the revisions of definitions would need to proceed along “smooth and natural ways,” meaning that any extension of the rails should be done only when required by the force of observation or argument. Which may mean, of course, that some such definitions, namely, those serving the field of metaphysics, would have to be unchangeable, if true. But the requirement of timelessness would itself be based on our findings, not imposed a priori.

The charge that this position is but a roundabout way of putting the old-fashioned nominalist case is a serious one, also. In order to do full justice to it, however, one would need to develop a theory of perceptual knowledge, one which would show that not all knowledge is propositional. Thus, although propositional knowledge, resting as it must on the knowledge of definitions, may appear to be ungrounded in reality, because such knowledge is itself grounded in perceptual awareness, which does not rest on knowledge of definitions, nominalism is avoided. The provision of the present position that requires that a definition correctly state the place some range of things, events, etc., occupies in a scheme of categories is tied to the additional provision that human beings can become aware of nature by way of their senses, that they gain evidence of existence by direct observation, and that it is by way of the rational organization—differentiation and integration—of what they are aware of that a rational scheme of categories is developed by them. The places carved in such a scheme are justified by a process of careful (scientific, ordinary, philosophical, etc.) reasoning, at least at their best or most accurate. And to say that some definition is correct or true is to say just that it would properly fit into such a scheme as is rationally developed by human beings. Anything beyond that—e.g., some alleged “real” final order of nature—is philosophical fantasy, so the inability of the present account to accommodate the desire for it cannot be regarded as a genuine shortcoming. Nominalism, on the other hand, does not enable us to secure the sort of direct ties to reality that would be made possible via a theory of perceptual knowledge which the present viewpoint presupposes. Of course, without the success of such a theory, the present conception of knowledge and definitions rests on shaky grounds.20

At this point it will be necessary to move on to a consideration of how the above may make possible the recovery of an adequate account of moral (and political) knowledge.


Earlier it was pointed out that a theory of knowledge would take account of the fact that although knowledge needs to involve firm awareness of reality, the sort of awareness of reality involved may depend on the aspect of reality in question. For example, it was noted that while definitions of concepts in metaphysics may have the character of necessary truths, definitions of concepts in astrophysics probably would not. At this point it will be necessary to go a bit further by offering what we could call an ontological hypothesis. The proposal is that nature is fundamentally multifaceted, multi-aspectival, or comprised of various modes of existence which are (at least from the point of view of present understanding) irreducibly unique.

What this proposal comes to is that existence seems not merely the duplication or multiplication of one kind of being into an endless number, but rather the presence of various kinds and types and modes of being. For example, we are aware of such drastically different types of existent things as time, music, objects, distances, animals, memories, edicts, commands, etc., etc. Notice that the list includes different types, not just different kinds. Reality seems clearly to confront us with a variety of ontological domains or spheres. At the same time, reality also seems integrated in terms of some general, what Aristotle called first, principles; e.g., contradictions are impossible in nature, regardless of which ontological domain is at issue. The principle of noncontradiction is, from all that we know, universal, transcending all particulars as well as different types or ontological domains.

On the other hand, there seems, also, to reign a pluralism atop this monism of basic principles, one which embodies the fact of the widest imaginable range of variety in what sorts of beings there are in reality. Things true of some sorts are clearly not of others— e.g., animals need nourishment, rocks don’t; melodies can be harmonized, weeks or days cannot. The facts involved in some aspects of reality are entirely absent in some others, as a matter of ontological distinctiveness. Memories may be faint but feet could not be; commands may be firm but distances could not. The point is that contrary to reductionist views reality is inherently diverse. That is, it is impossible to account for the facts about all aspects of reality by reference to just one distinct aspect, e.g., the physical or mental or numerical, except, perhaps, by allowing for a very extensive evolutionary process. Reductionist theorists tend, on the whole, to be promissory. Furthermore, even if in some sense all things might be given an account of simply in terms of some one kind of thing, that is not what is relevant nor is it the way human beings, including those who propose that idea, make sense of reality.

While in detailed metaphysical discussions the above ontological hypothesis would require extensive support, for present purposes it will have to be taken as merely a plausible backdrop for certain epistemological considerations. Given what has already been said about the nature of knowledge and definitions, the ontological hypothesis suggested above may shed light on whether moral (or political) knowledge is possible. To wit, if there are irreducible differences between distinct ontological realms, and if we can obtain knowledge of facts within distinct ontological realms, the character of knowledge involved in knowing these facts could very well reflect the differences at issue. Certainly, the idea that what we know—our objects of knowledge—could have a decisive impact on the character (or sense) of the knowledge involved is not implausible. Knowing the time of day could very well be knowledge of considerable distinction compared to knowing the name of all the kinds of alloys or compared to knowing what will most readily upset Susan’s grandmother. The sort of things we should use to test these different cases of alleged knowing would themselves differ considerably.

Now it is not merely meant here that knowing one thing is different from knowing another, something that is plain enough, of course. Rather, the possibility is being advanced that when we in fact know some things in a given ontological domain—e.g., a song in the musical or a rock in the physical or, again, someone’s recollections in the mental—the character of the requirements to be met so as to establish that we in fact have (or our justification of) knowledge will differ. Thus to test whether someone knows a song, one requires that at least some of the crucial passages can be at least doodled by the person making the knowledge claim. Whereas to test whether one knows what limestone is would appear to require the correct listing of chemical components which comprise the material in question. To know, what someone’s recollections of last summer at the beach are would have to be tested, in turn, e.g., by comparing that person’s reports with what he’s said before when asked. There are, of course, common aspects to these tests— awareness needs to be demonstrated. But the awareness will take a different form in the different cases, calling to the fore different skills and utilizing different faculties—e.g., good ear, acute observation and learning, familiarity with an individual. In general, there can be an indefinite number of types of knowing, with different expectations associated with each type. Knowing the past, the present, the future, and the possible can involve very different sorts of evidence, argument, and performance.

Accordingly, instead of the familiar distinction between knowledge of is and knowledge of ought, a much more diverse conception of knowledge is being suggested here. The idea is that, probably, parallel to each ontological domain a mode or type of knowledge has emerged. For example, mathematical knowledge probably has its unique features, distinct from biological knowledge. In turn, if there is a genuine, irreducible moral domain—either fully or partially merged with the political, or, perhaps autonomous—there is, probably, also a unique mode of knowledge that corresponds to it.

The reason for the hesitancy in putting these points forth is that one would not be able to argue for them in any ordinary way, certainly not in the way familiar in the sort of epistemological and meta-ethical works as Lehrer’s, Chisholm’s, and von Wright’s and Hare’s, respectively.21 But there is an argument here, and it takes the form of an attempt to make the best sense of a range of phenomena, namely, that range that constitutes the interrelated concerns involved in trying to understand morality. In short the general framework is being spelled out in such a way that we can accept that moral knowledge is possible without jeopardizing evident differences between purported moral knowledge and other knowledge the denial of which would be unreasonable. If the model of a conception or definition of knowledge Lehrer and others embrace is indeed prejudiced, as argued, and if it is reasonable to open up our conception of definitions, including the nature of something, then, with the possibility of the multifaceted ideas of both reality and knowledge, there could be ample room for moral knowledge despite the type of arguments that moral skeptics present. Of course, this last point would benefit from some illustration.

For example, the sort of argument presented in the so-called naturalistic fallacy objection to moral knowledge trades heavily on a narrow conception of what it means that the nature of goodness is X. The open question argument assumes that when something has X as its nature, it must be inconceivable that anything without X could nevertheless be best regarded as being an X. But if thing is X—has as its nature X—that thing in some perfectly respectable manner could be imagined not to be X. Clint Eastwood is a macho character in the movies but he is easily imaginable as not being such a character in some movies. That horse down the field can be imagined to be but a statue of a horse, or perhaps a movie image of one.

Earlier it was noted that any logically possible counterexample need not serve to place doubt in our definition of knowledge, with Gettier type cases in mind. In connection with a complex field such as moral knowledge, the Gettier type counterexamples can be overwhelming. (Judith Jarvis Thomson is an artist in producing them, as is Robert Nozick.) But if definitions can be right without conforming to that strict conception of what they must be, as spelled out in so much of recent epistemology, then the definition of goodness and of moral goodness might not be so insurmountable a task as it has seemed.

Apart from that very general point, which in a broad sense illustrates how moral knowledge might be possible—i.e., by removing some familiar obstacles from its path to success—it would also help considering some specific example of judging some deed as right or wrong. But this is equally difficult, if not more so, as making the general point. The familiar desert-island or raft-boat cases, or even the fat man on the San Francisco bridge who should or should not be used to save the life of many others in a run away cable car, are inadequate because their context is hopelessly sketchy. They establish nothing much, whichever way they go. It is no wonder that the intuitionist approach appears to suit those cases best—at least the subconscious mind has developed the needed reflexes, in most cases, to come to grips with some such cases. But the complexity and number of the facts that are filled in for purposes of getting a reasonably complete picture—a “real life” situation—are duplicable usually only in actual case histories or very well constructed novels, plays, songs, etc.

There is, moreover, something special about moral knowledge which is perhaps not there in any other sort, namely, that it is severely self-referential and tempting of self-deception. The stake any person has in particular moral knowledge is considerable— provided there is indeed a genuine, bona fide moral realm of human life. Such a realm concerns the very quality of a person’s actions and, ultimately the person’s life—himself or herself. That alone poses problems, because there is special interest, tempting everyone, to plead the case for some way of conceiving of goodness so that one enhances it more than not. In other fields of human understanding there is some of that temptation, too, but mostly confined to special competence. The moral dimensions of being lazy on the job, stealing someone else’s ideas or property at work, faking some task a bit, even being merely negligent with rather meager impact—the dimensions of these are narrow, although sometimes intensely felt. A scientist who betrays the standards of the profession is morally guilty. Yet mistakes, even failures, in special fields do not seem to cast general aspersions on the person. The answer to the sort of question, “What, in general, is to be morally good?” in contrast unearths fundamentals, matters that span the details of one’s life.

These last points alone prove nothing. They suggest, however, that lack of wide agreement, widespread disagreement, as to what is morally right or good need not be due to the truth of moral skepticism. They also aid in appreciating the impact of the wrong- headed model of moral knowledge imported from general epistemology and the model of definitional knowledge. With many people having a special interest in watering down any moral theory—some even having a professional or sporting interest in debunking its very possibility—the interest in constructing possible counterexamples may be considerable.

Furthermore, moral knowledge could indeed be extremely diverse in itself, making room for variations based on all sorts of factors. We should perhaps think again, of morality as the science, and of ethics or applied morality or praxis, as the resulting engineering, with the important addition that a lot of the science is inadequate, even fake, and much of the engineering ill guided and even misdirected. Even apart from these features, moral knowledge at its best will be highly individuated in some practical situations. Weighing the importance of things for any person, even with a very general common framework available and shared with others, is difficult. Against this it should, however, be recalled that within all the variations of what is morally right, there runs the widespread awareness of some commonality, one that prompts the question in such categorical ways as “What, in general, ought a human being be like?,” “What is the morally good life?” and “How can I lead my life in a virtuous, proper, self-respecting manner?” Even convinced skeptics transform such questions rather than treat them as on par with those of palmistry or voodoo?22

Given, then, that moral knowledge is not precluded by any general “is/ought” problem, given that defining moral goodness could be managed, and given that something else besides its impossibility explains the problems found with moral knowledge, where does this leave us?

One promising approach appears to be intuitionism in its best form, as spelled out by Mark Platt,23 for example, where we are told that there simply are clear cases of moral knowledge no one can reasonably dispute. I will not spend time disputing this doctrine but merely state my dissatisfaction with it, which is based on the fact that the integration of this intuitionism with the rest of epistemology is problematic. It leaves as a problem, I believe, the coherence of the relationship between moral knowledge and knowledge of other matters. It also makes it difficult if not entirely hopeless to try to resolve conflicting intuitions. But whether this is indeed the way that theory stands cannot be examined here.

It may now be suggested, instead, that the distinctive moral domain emerges when we reflect on the fact of human life. First of all, the broad domain of value appears with the emergence of life per se.24 When we consider the nature of life as self-generated behavior that can foster either sustenance or demise, the category of something being of value to the living thing emerges and acquires clear significance and meaning. It is for the living that something is of value or disvalue, since living things alone are faced with the constant possibility of extinction.

Now even before proceeding, we can see the way that a defective epistemological framework invites devastating challenges to the above. Consider some of the objections Robert Nozick poses to Ayn Rand’s efforts to develop a similar theme.25 Nozick constructs some rather wild counterexamples, involving computers and other sorts of indestructible beings and imagines that these can benefit from something or experience harm, thus suggesting that the points made above about life are certainly not necessarily true. Therefore, Nozick concludes, the view that the basis of value is the existence of life is unfounded. If, however, we reject the epistemology presupposed in such attempted refutations, a lot more than cleverly imagined counterexamples would be needed to create serious doubt about the view.26 It is clearly a plausible view and on reflection it makes very good sense. We in fact judge things to be good and bad, in connection with not just human life but life in general, based on whether something flourishes or is aided in its flourishing. Botanists, zoologists, and physicians clearly use the sort of life at issue as their standard for judging the goodness or value of particular cases under consideration—e.g., some tree in a park, some animal in a zoo, or some patient on the operating table. And the whole array of judgments of good and bad seem to be grounded on considerations of whether something contributes to, rather than hinders, the life that is of concern to the judge.

When we turn to moral values, it seems evident, again, that a special form of life gives rise to the topic. In general, we see that being as fully human and as completely consistent within ourselves as the case makes possible (i.e., fully integrated) is in our own power, and exercising our power in the various ways we do is just what gives rise to moral evaluation.27 In the case of human life, whether it goes well or badly—never mind for the present by what standard—is an issue of personal achievement or failure, respectively, at least in the bulk of cases. At least where matters over which persons can exercise control are concerned, their actions and life, in general, are exposed to moral scrutiny. This category of goodness, namely, moral goodness (or evil) emerges in conjunction with the apparently distinctive human capacity for freedom of choice, for then there can arise the issue of whether a person has chosen to act in accordance with a standard appropriate for leading a human life. (It should be noted here that while all this may appear to be wedded to some absolutist, universalist ethical perspective, once the metaphysical status of individual human beings is recognized—namely, that “human being” exists only insofar as there are individuals whose place is best fixed in a rational scheme of classification by the definition of that concept—it is possible to show that the standard is not so abstract as it may appear.) It is the fact (if it is one, which I will not discuss here) that human beings are self-determined that introduces a distinctive ontological domain wherein the standard of knowledge—of what something must be to count as bona fide knowledge—could be different from what it is in others (e.g., physics, biology, economics.)3° An entity that can cause at least some of its action, that presupposes a degree of causal agency, really appears to warrant differentiation from others. Here again, however, the differentiation need not be understood as some kind of absolutely fixed, timeless category—the definition of the concept “human being” need not involve some necessary truth in the formalist sense. This means that it is not necessarily ruled out that some heretofore nonrational animals will join the category of beings properly designated as human or that such an animal will join that category in some significant respect and thus possibly acquire rights.31

Now we can consider that the category of moral knowledge may involve what is the case concerning actions, institutions, decisions, etc., pertaining to human living and see what form such knowledge might take. If morality involved standards for conducting a human life, and if human beings are largely free to choose what they will do, we might have for the facts of morality a set of hypothetical imperatives rather than categorical truths. If one is to live one’s (human) life, then one must, in general, carry on appropriately (consistent with what and who one is). The content of this hypothetical will depend on facts pertaining to human beings as such, i.e., human nature, and facts about the individual person in question. There is no room for a categorical moral judgment here because whether one is to live a human life is, in the last analysis, a matter of choice. Granted, such a fundamental choice, if it is to be coherent, meaningful, and capable of implementation, is itself limited to either living a human life—for that is what a human being is free to do in view of being a human being—or not living a human life, that is, extinction, ceasing to live at all. The former choice, if made, commits a person (i.e., oneself) to some very general principles of conduct, based on what it is to be a human being, but leaves much to the determination of the special circumstances of one’s particular sort of life—age, culture, sex, talents, capacities, opportunities, etc. The hypothetical imperative of “If one chooses to live, one must, in general, live as it is fitting for a human being” can now be filled out a bit by noting that human beings are, as such, free and rational animals. Their freedom and reason are not, in fact, separate but related aspects—it is in their capacity to reason (to reflect and act by judgments, principles, theories that they can evaluate as sound or unsound) that persons embody their freedom of choice.32 The proper standards for human life, then, must ultimately be related to the fact that each person is free/rational and a biological entity whose life is lived well (as a matter of his or her choices, apart from accidents that cannot be controlled) insofar as it does justice to the fact of an essential dependence on reasoning and a natural requirement for biological health.

We will not go beyond these sketchy remarks to formulate a complete moral theory because the present discussion needs now to be extended into the public normative realm, namely politics, and because elsewhere the details of the above perspective have been developed to a considerable extent. Certainly more could be said about moral knowledge and its content, but there is no room for this here.33


Much of Western political theory has rested on the conclusion that moral knowledge is impossible.34 Defenses of liberal democracies have tended to stress considerations of general welfare and diversity, not the moral worth of the system or its capacity to give rise to morally good citizens and morally worthwhile institutions. The common normative theme of such societies has been the political equality of all persons, a theme that is then often extended to imply each person’s (at least prima facie) moral equality. Without the prospect of moral knowledge the only, albeit very meager, moral message that is plausible is that we are all of equal worth.

It seems clear that many Western thinkers, from the time of David Hume to our own, have equated the possibility of moral knowledge with the entitlement to impose conformity to that knowledge on others. There is ultimately no logic in this extension, but a certain prevailing view of the human good has lent support to it. Both utilitarians and Marxists are intrincisists with respect to the human good—for the former, the greatest happiness of the greatest number; for the latter, the fullest realization of human nature constitutes the human good—and the right course of conduct amounts to furthering the probability of reaching this human good. Thus, if by chance making another behave in certain ways can enhance the likelihood of the realization of this good, then we must institute perhaps even coercive means to ensure performance of various kinds of appropriate behavior. The best defense against this threat to moral independence and civil liberties seems to be to maintain that no one can know what the human good is (or to maintain that it is relative).35

The main problem with this approach is that it accepts a false dichotomy between two equally flawed views. It is false that we cannot know what is morally right, but it is equally false that knowing it entitles one to force another to behave accordingly. This is evident enough to most, whatever a philosopher says. It is clearly enough recognized that moral knowledge does not by itself entitle us to force another to act according to it. We know in the case of many people that it is morally bad that they are gluttonous and morally good that they practice temperance, but we also know that this alone does not entitle us to force such a person to reform.

The conception of the good human life implicit in the points raised earlier is different from the intrincisist view. It is not an end-state that is to be achieved by living a morally worthwhile life; it is such a life itself that is the good, and it is choosing to live it that makes it both possible and morally praiseworthy. Making another live a morally good life is, in turn, impossible, for the morally good life requires free choice.36 In contrast, the current concern with the public interest or common good, as evident in virtually all discussions of morality and public policy—discussions that too often avoid reference to personal virtues or character— invokes the end-state approach. This assumes, largely on the basis of intuitions or considered moral judgments, so called, that certain states are morally imperative—equality, harmony, family life, chastity, political stability, employment, safety, security. Then means are discerned for reaching some combination of such states, regardless of how these means bear on the personal and moral independence or autonomy of individual members of society. Such a conception of the public or common good is questionable and advanced largely independently of the rest of philosophy.

The metaphysical necessity of freedom, or self-determination, for living a morally worthy life based on the rest of the points raised earlier in this essay, suggests different ways of conceiving of a good community life. Any community’s basic standards of human intercourse must incorporate a recognition of the indispensability of a sphere of moral responsibility for each person. Persons need room to realize themselves through themselves—that is, need a sphere of authority for reaching the highest possibility qua the human beings and the individual they are. This abstract point supports a feature of the Western liberal political tradition, namely, that the life of the individual is the center of value—i.e., the common political good—from which the various institutions of society gain their importance and according to which the legal framework of a society ought to be devised, reformed, maintained, protected, and so forth. But when we consider the matter along lines suggested here, we see that a system of political and economic liberty that leaves it to individual and cooperative effort morally to secure the diverse goals of human existence, is not just a next- best system based on skeptical humility. It is a rational consequence—derived (but not deduced) from our knowledge of man qua man—of the fact that each person has a moral task and none other can perform it for him.

There are some disappointments, of course, with this perspective. For one, nothing so robust as the total emancipation of humankind or society is demanded in the political arena, because human beings as such are always free to fail at their moral tasks of becoming what they should become as human beings and the individuals they are. Furthermore, it will not be possible, as the liberal Zeitgeist seems to desire, to proclaim all lifestyles morally worthy, however politically proper and indispensable it is that many such lifestyles (namely all that do not involve infringement of others’ right to liberty) receive the full protection of the law.

What some supporters of Western values have argued is quite true, and furthermore the truth is not a result of our inability to know better but our ability to know quite well. Liberty is the highest political value, the deciding principle within the public domain. But it is so because it is right for human beings in one another’s company to refrain from obstructing others’ basic task, namely, to live in dignity, to make for themselves a morally worthwhile life.

There is nothing philosophically odd about this point of view, nothing in a reasonable theory of knowledge that cannot support it as a viable competing alternative. Furthermore, the approach sketched is comparable to the approaches of the great political theorists, excepting that the conception of the good life for man is not idealistic to the point, as under a prominent interpretation of Plato (and even of Aristotle), of being an impossible dream. Thus, a task urged upon many Western thinkers may be met by way of a detailed development of the present approach, namely, to formulate a comprehensive alternative to the seductive holism of Marxist ideology without, however, encouraging totalitarianism. Marxism, unlike the position sketched here, subsumes all normative considerations under political economy, whereas the present view leaves to individuals (in their various positions—alone, in families, with friends, with colleagues, etc.) the task of moral excellence and to political organizations the task of enabling the individual, in the face of dangers from his fellows, to carry out this task in peace. This limited conception of government is in need of moral justification, and the present approach would appear to be a very hopeful try at that.


The possibility of objective knowledge that is not reducible to knowledge of one or another kind of stuff (ideas, sense data) makes possible the introduction of a means by which the “is/ought” gap may be avoided. The possibility of a definition of man (or of the concept “human being”) makes possible a theory of the human moral good that avoids relativism. The fact that this moral good must be of a very general type, with application to all human beings, suggests that enormous variety will exist at the numerous less general normative levels—e.g., what is morally right conduct for senators in 350 B.C. Athens, what is a proper code for students in a society where there exists compulsory education, and what scientists should do with reference to globally destructive weapons, how parents in Central Africa should rear their children, etc. That the nature of man involves the freedom to choose one’s conduct given one’s situation, and the responsibility to do this well, requires that others abstain from coercing one to act in certain ways—i.e., that they respect political or civil and economic liberty.

     We can only evaluate the case for objectivity in the domain of norms when we compare it to the case for relativism or skepticism in this domain. In this respect, we need to be like detectives or scientists and engage in comparative analysis. The approach advanced here in outline seems to account for the various needs we have in the area of norms and it seems, also, to take account of some of the attractive aspects of different theories. The skeptic is concerned not to cut off further inquiry, not to get boxed into agreeing to what is not so. The nominalist wants to account for creativity and evolution. The moral and political relativist wishes to make room for the enormous diversity in human life, diversity of which it would be odd to claim that it signifies moral evil.

     The position I have sketched here begins to achieve the task of assimilating these various features. Of course, it will not assimilate all features. It is better than others, and those who find themselves in fundamental disagreement with it cannot expect that it will yield to their position. In this it satisfies the need we have for some stability, for standards for distinguishing right from wrong.

     Some controversial implications of this approach have now been indicated. Indeed, the present position, with its contextualist epistemology, its ontological pluralism, and its individualist ethics does lead to a political system that favors individual rights over collective goals, the present over the future, and liberty over equality—to give just a few important results. Unfortunately, a great deal had to be accomplished in a short space, and giving the position full evaluation at yet another comparative level requires further work.37


1 By “success” I mean the achievement of prominence in the philosophical community. For some reflections on why such variation of success has occurred, see Leo Strauss, Natural Right and History (Chicago: University of Chicago Press, 1953).

2 In some contexts these terms are interchangeable. But more strictly considered, the essence of something is what is distinctive about it, while its nature is what it requires to be. For example, if human beings are by nature rational animals, they are essentially rational. I am here concerned with the nature of something, not merely its essence.

3 See John Tienson, “On Analyzing Knowledge,” Philosophical Studies, 25 (May 1974): 289-93 and Robert Nozick, Philosophical Explanations (Cambridge: Harvard University Press, 1981), chapter 3.

4 Keith Lehrer, Knowledge (Oxford: Oxford University Press, 1974), p. 6.

5 Ibid., p. 7.

6 Ibid., p. 6.

7 As noted, Lehrer is not alone. Nor is he alone in drawing skeptical conclusion from the ensuing analysis. See, for example, Peter Unger, Ignorance (Oxford: Oxford University Press, 1975).

8 See, for example, Douglas Rasmussen, “Logical Possibility, Iron Bars, and Necessary Truths,” New Scholasticism, 51 (Winter 1977): 117— 22.

9 J. L. Austin, Philosophical Papers (Oxford: Oxford University Press,

10 C.f., Edmund Gettier, Jr., “Is Justified True Belief Knowledge?” Analysis, 23 (1963): 121—123.

11 Gilbert Harman, Thought (Princeton: Princeton University Press, 1973), p. 145.

12 Barry Stroud, “Wittgenstein and Logical Necessity,” in G. Pitcher,
ed., Wittgenstein (Garden City, N.Y.: Anchor Books, 1966), p. 496.

13 Ibid.

14 This scheme need not be a final, closed one. For the full view of
definitions invoked here, see Ayn Rand, Introduction to Objectivist Epistemology (New York: Mentor Books, 1979). But see, Larry Briskman, “Skinnerism and Pseudo-Science,” Philosophy of the Social Sciences, 9 (1979): 81—103, and see, also, Larry Briskman, “Essentialism Without Inner Natures?” Philosophy of the Social Sciences (forthcoming) where our initial reflections are severely criticized from a Popperian point of view. Many of the points we make here can serve as answers to this criticism.

15 We have in mind the point that when there is knowledge, as it is familiar to us in our daily experiences, it is had by someone, somewhere, at some time.

16 There is also a distinction to be made between counterexamples and borderline cases. The view that if there are borderline cases, then we have no justification for adopting a definition as correct assumes, usually, that definitions are necessary truths. As to the character of successful counterexamples, see Kenneth S. Lucey, “Counterexamples and Borderline Cases,” The Personalist, 57 (1976): 351—355. Arthur W. Collins, “Philosophical Imagination,” American Philosophical Quarterly, 4 (January 1967): 49—56.

17 For example, in the domain of metaphysics, it would turn out that concepts are defined so that a valid concept must be defined by a necessary truth, mainly because such a concept will serve in a proposition that refers to everything, everywhere, at all times, so changes could spell having to give up the truth of the proposition. The same may hold for other fundamental fields, e.g., mathematics and logic. Our concepts of the atom and of cells, however, have undergone considerable change. But it is not justified to believe that such change entails that the earlier understanding was wrong.

18 Even a neo-Platonist such as Leo Strauss sees the matter along these lines, when he says:

Socrates started in his understanding of the natures of things from the opinions about their natures. For every opinion is based on some awareness, on some perception with the mind’s eye, of something. Socrates implied that disregarding the opinions about the natures of things would amount to abandoning the most important access to reality which we have, or the most important vestiges of the truth which are within our reach. (Natural Right and History, pp. 123- 124.)

The problem with Socrates seems to me to have been his extreme reliance on a method of comparing and contrasting such opinions. This method seems to have been based, in part, on the acceptance of the participants’ goodwill. Still, this engagement of the opinions of attentive people seems justified because it is first of all through the immediate encounter of reality that we sort things out and begin to take more interest in some things than in others. Thus, Plato has Socrates consider many problems by embarking on the construction of solutions with the aid of loyal pupils, whereas, observes Leo Strauss in The City and Man (Chicago: University of Chicago Press, 1964), “in the conversation between Socrates and Thrasymachus, justice is treated in a bantering and hence unjust manner (since) Thrasymachus . . . does not take seriously the virtue under discussion” (p. 85).

19 Descartes exemplifies this unfortunate approach, in my view, and his influence seems to be responsible, in large measure, for subsequent trends in epistemology. It is Kant’s continuation of the trend set by Descartes that provoked Nietzsche to lament the philosopher’s approach to truth. See Freidrich Nietzsche, Beyond Good and Evil, trans. Walter Kaufman (New York: Vintage Books, 1966).

20 See David Kelly, Evidence of the Senses (Princeton University: Unpublished doctoral dissertation, 1978). See also Thomas A. Russman, “Selective Perception,” Reason Papers, No. 7 (Spring 1981): 21—32.

21 It is the formalism of these philosophers, in their discussion of their respective fields, that I am calling attention to here.

22 Among others, B. F. Skinner feels the need to do this in Science and Human Behavior (New York: Macmillan Publishing Co., 1953), p. 429.

23 Mark Platt, Ways of Meaning (London: Routledge & Kegan Paul, 1979), chapter 10, and Reference, Truth and Reality (London: Routledge & Kegan Paul, 1980), chapter 4.

24 Only what is extinguishable can be the beneficiary of something— a process, thing, prospect---that might prolong its existence. Apart from the life of some organism, nothing else is extinguishable, only transformable, so that only life can experience what is of value to it, of benefit to it. A somewhat more elaborate way of putting this can be found in Ayn Rand, The Virtue of Selfishness (New York: Signer Books, 1964), p. 20ff. The point is developed fully in Douglas Den Uyl and Douglas Rasmussen, “Nozick on the Randian Argument,” The Personalist, 59 (April 1978): 108—205.  In the present discussion we wish merely to indicate that value or goodness or benefit makes its appearance with the emergence of the ontological domain of living things for which something can be of value, good, of benefit, etc.

25 Robert Nozick, “On the Randian Argument,” The Personalist, 52 (Spring 1971): 282-304. But see, Den Uyl and Rasmussen, “Nozick on the Randian Argument.”

26 Here is where an account of “logically possible” is crucial, since mere imaginability is often accepted as showing logical possibility, so that when some counterexample is taken to be logically possible, it is argued that from a logical point of view a conclusion cannot be maintained. See my “A Note on Conceivability and Logical Possibility,” Kinesis, 2 (Fall 1969): 39—42.

27 I discuss this in more detail in my A Primer on Ethics (Santa Barbara, Calif.: Reason Press, 1982).

28 In other words, given that human beings are necessarily individuals existing sometimes and somewhere, the general principles of morality and politics will probably apply to them quite differently, depending on the facts they face in their lives.

29 See Roger W. Sperry, “Changing Concepts of Consciousness and Free Will,” Perspectives in Biology and Medicine, (August 1976): 9—19.

30 This position mirrors, in the metaphysical sphere, the epistemological pluralism we find in Stephen Toulmin, The Uses of Argument (Cambridge: Cambridge University Press, 1958). We mean just that in various contexts the criterion of knowledge will be different from what it is in others.

31 Many philosophers argue for animal rights today. My acceptance of the fact that it may not be ruled out that animals will have rights is not the same as endorsing their views. (For there to be animal rights, animals would have to be able to choose between exercising or not exercising their rights, something that is not now in evidence.)

32 I develop this point in my The Pseudo-Science of B. F. Skinner (New Rochelle, New York: Arlington House, 1974).

33 I say a bit more in my “A Reconsideration of Natural Rights Theory,” American Philosophical Quarterly, 19 (January 1982): 61—72.

34 For example, see Karl Popper, The Open Society and Its Enemies (Princeton, N.J.: Princeton University Press, 1963).

35 For example, Larry Briskman, “Skinnerism and Pseudo-Science,” Philosophy of the Social Sciences, 2 (May 1979): 81—103. Cf., Ranford Bambrough, Moral Skepticism and Moral Knowledge (Atlantic High, N.J.: Humanities Press, 1979).
36 For a more developed discussion, see Douglas Den Uyl, “Freedom and Virtue,” in T. R. Machan, ed., The Libertarian Reader (Totowa, N.J.: Rowman and Littlefield, 1982).

37 I wish to thank the Earhart Foundation for support for some of the work on this paper. My colleague Randall Dipert and the editor of The Review of Metaphysics have given generous help toward the improvement of an earlier draft, as has Marty Zupan.

Return to Tibor R. Machan Page