The Argument for Metaphysical Universals

“It is Aristotle who identified the fact that only concretes exist. But Aristotle held that definitions refer to metaphysical essences, which exist in concretes as a special element or formative power… Aristotle regarded “essence” as metaphysical; Objectivism regards it as epistemological.”

“Thus the essence of a concept is determined contextually and may be altered with the growth of man’s knowledge. The metaphysical referent of man’s concepts is not a special, separate metaphysical essence, but the total of the facts of reality he has observed

“I was discussing the issue of concepts with a Jesuit, who philosophically was a Thomist. He was holding to the Aristotelian position that concepts refer to an essence in concretes. And he specifically referred to ‘manness’ in man and ‘roseness’ in roses. I was arguing with him that there is no such thing, and that these names refer merely to an organization of concretes, that this is our way of organizing concretes.”

Introduction to Objectivist Epistemology (bold emphasis mine)

“Epistemic universals”

Rand denies metaphysical universals quite explicitly, as quoted above. She believes everything in reality is concrete and particular, that there is in reality no “manness” in man which applies universally for all men at all times, but rather the concept “man” is merely man’s way of organizing the concretes seen around him into a mental grouping.

“The metaphysical referent of man’s concepts is … the facts of reality he has observed.”

If one holds that concepts are only “universal” over the total set of of one’s prior, concrete observations, then this is not the universal set! This isn’t guaranteed by any metaphysical principle to hold at all times and for all instances in reality. Concepts in this view aren’t describing something that holds abstractly in reality, they are just describing something that holds abstractly over the particular, delimited set of observations which one has accumulated thus far.

If “universals” are merely referring to sets of observed particulars, then one cannot interpret anything observed, predict the future, or classify anything new, when nothing in general about reality can be referred to. The “man” classified today might have nothing to do with the next “man” observed. The ball observed in one moment tells one nothing about what might be observed in the next moment. Any particular, any moment yet to be observed, nothing can be said about it, because the classifications are all retrospective, they only refer to the particulars already observed.

The “epistemic universal” of “length” one invents today can say nothing about the “length” observed tomorrow, because no necessary connection is being induced, nothing general about reality itself, it is just the cataloging of regularities in experience. They are just retrospective statistical observations – the moment one starts talking about length – every property of length in all places and all times – then one is talking about a universal property out in reality, a metaphysical universal, which is exactly what has been rejected.

No inference can be extended to particulars outside of the cherry-picked set of concretes previously observed. If a concept “stands for” an unlimited range of things abstractly, but concretely it only refers to some particular set of items already identified, then there is no way to know if the abstraction actually does apply to the full range of things that it stands for.

One can define a category of “winged things” which is open-ended, and therefore includes all winged things yet to be observed. Obviously any new instance added to the set will have wings, but nothing else can be said of it besides that. Without such a thing as a natural class, then what is formed is merely a nominal category, in other words the category is merely analytical, and the only thing that can be inferred from classifying something as a “winged thing” is that it has wings. Which is of course useless.

If there is no natural kind backing the concept, then there’s no justification for inferring anything beyond what has already been defined. If on the other hand concepts are identifying a natural kind, then there’s a necessary connection between all particulars in the set, from which one can justifiably infer things like “any new particulars added to the set will behave as the rest of the set”.

If one holds that “any new particulars added to the set will behave as the rest of the set”, then one is apparently identifying a universal in reality. It functions as a universal, and abstractly identifies something in reality that is timeless and essential, something where instances at all times and in all cases will behave in that same way. If an abstraction is be extended across all instances at all times, and out into reality (in the sense that it will predict the future behavior of things in reality), then the abstraction is something that is metaphysical and universal. A nominalist is someone who rejects that any such thing is metaphysically possible or epistemologically justifiable.

Universals which “hold true” but do not “exist”

Rand believes everything in reality is concrete, that, in reality, there is “no such thing” as the universal “manness” which ties together all concrete men, at all times and in all places. This “manness” is rather our organization of concrete men.

She claims that, by properly organizing concrete men, one can thus arrive at a universal “manness” which will hold true for all concrete men, at all times and in all places.

So the universal does exist mentally but not in reality? Does it “hold true” in reality, and just doesn’t “exist” in reality? There is this odd reluctance to grant the existence of something “in reality”.

Dual aspect metaphysics grants this idea of an “abstract reality”. Some abstraction which holds true in reality, therefore is real. It gives a kind of reifying existence and power to the abstraction, the abstraction is what is metaphysically making it hold true, as opposed to something else making it hold true and the abstraction merely epistemologically “recognizing” that the truth is holding, presumably for some other reason.

It’s kind of an odd question- what is the real thing which is making this universal hold true? There must be something with the force of reality which is making this truth hold- what is that force? Where does that force come from?

Rand asserts that there are no abstractions with this power: only concretes are “really real”. But even some given concrete has to have some abstract nature. Is the material of the concrete supposed to be powering the nature of the concrete? It doesn’t really make any sense when thought about clearly. Only the dual aspect perspective, a la Aristotle’s hylomorphic compounds, actually makes any sense.

Apparently Rand’s perspective is that one cannot say why, but things just “happen” to work universally. That’s just the way the concretes behave- but they don’t behave that way because of some abstract principle of their nature. That form or principle is just a “way man describes” what matter is doing, it only exists in the mind, not in reality itself.

It is bizarre to say that and also hold that induction is possible, as in McCaskey’s article, where he insists that it is possible to have 100% certainty about regularities despite there being no principle of uniformity. How can one have 100% certainty that a regularity will hold, if one denies the reality of some principle to it? There is no way to make a valid inference from any number of observations of a behavior to a universal rule of the behavior. What is to say it won’t change, if it is not a real aspect of the thing’s nature?

McCaskey for example claims:

“If you have good guidelines and follow them, you can be certain that someone absolutely cannot contract cholera unless exposed to the bacterium Vibrio cholerae, certain that all men are mortal, certain that the angles of all planar triangles sum to 180°, and certain that 2+3=5.”

Well no, under this system, none of these is certain. No conclusion science has ever achieved can be described as true, or knowledge, or certain. They simply happen to be true under the concretes previously observed, and one predicts it will continue to be so.

Yes, even with math. Is it certain that 2+3=5? At all times and in all places, universally? How? That may have held up under previous observations, and one may predict that it will continue under similar circumstances, but all of one’s predictions are unjustified and unreliable; one hasn’t observed every single instance that has ever occurred.

From the article: “It’s not that you must presume uniformity in order to classify. It’s that you classify to find uniformities.”

The whole problem with this is that one hasn’t “found” any more uniformity than one had to begin with! One is left in exactly the same position as he agreed with earlier in the article: “The Scholastics lamented (rightly) that unless you had surveyed all magnets or all animals, the inference was not certain”

“Certainty” without proof

A fall back here is to argue that concepts are not universal, but that one can still have a kind of “certainty” which could be mistaken, that it is still “knowledge” which one can hold beyond a reasonable doubt.

If something hasn’t been proven to be true then there is a reasonable doubt that it could be different at some other time or place. After all, it has been asserted that one cannot make justifiable claims for something at all times and all places.

If one is making a universal claim about something at all times and all places, and holds that such claims are invalid, cannot be justified, then how in the world can one feel certainty about them? It has been asserted that one can’t hold such general claims as true, certain, knowledge.

Universal claims are either justified or unjustified. One must choose. If they are unjustified then one cannot claim “certainty” and “the impossibility of doubt”. If universal claims are justifiable, and a given one is proven, then one can claim certainty and the impossibility of doubt.

Either the claim “2+3=5” is unjustified and therefore fallible, or else it is justified and therefore infallible.

It makes absolutely no sense to declare that some claim is unjustifiable, but also true beyond a reasonable doubt.


If one denies the existence of universals metaphysically, then there’s no reason to believe that an abstraction will extend beyond the range of the small set of previously observed concretes to which it currently refers (and certainly not to believe that one has knowledge or certainty about it). In that case these “universal” epistemological abstractions do not provide knowledge, one cannot have certainty about them – and indeed the opinion of a nominalist is that the use of or belief in such “epistemological universals” is foolish and counter-productive, after all, what’s the point in having or believing in some “timeless essential” if it’s not referring to something that is actually timeless and essential in reality? These universal abstractions are actually false and misleading, they distort the view of reality since there are no such things. There are only retrospective categories of reference.

Calling such epistemic categories “concepts” or “universals” is mistaken. None of the positive results that Ayn Rand tries to claim follow, like the ability to have conceptual knowledge, or certainty about reality, or the validity of induction. None of this is really consistent with this view; one is a skeptic about any general statement about reality, a nominalist who believes in categories of convenience, and the epistemic standard (and thus, necessarily, the moral and political standard) is subjective and pragmatic. There are plenty of people who own up to holding exactly this view, nominalists of all kinds own up to this and wear it proudly, declaring that all that is possible to man are pragmatically guided categories and statistical correlations, and that belief in concepts, in universals that actually hold in reality, is akin to a religious fantasy from which one must break away.

If on the other hand one is not truly a skeptic about reality, if deep down there is a belief that it is possible to justifiably know things that are necessary and certain and universal, then there must be a conversation about the metaphysics of universals. Either way there’s an inconsistency in Ayn Rand’s thinking, and one should be clear and honest with themselves on exactly where one stands.

One must choose a side. Either there are universals which actually hold in reality, or else there is no such thing. If there is nothing in reality which holds universally, then one cannot have knowledge of things which hold universally. It is not possible. One either needs to own up to one’s metaphysical stance epistemologically, or own up to one’s epistemological stance metaphysically. It cannot be held both ways. The concept of metaphysical universality cannot be stolen in epistemology while denying it in metaphysics – not if one is being honest with one’s self. Either one has a merely pragmatic stance (i.e. holding this concept as if it were a universal, even if there are no such things, since that seems to work well) in which case one ought to own up to that epistemologically as a nominalist, or else one does believe that universal knowledge is possible but is operating on a stolen metaphysical premise, in which case one ought to own up to that metaphysically as an intrinsicist.