1. A theory of universals is traditionally supposed to answer two questions: first, what makes generic identity possible across specific difference (e.g., what makes red horses and brown horses both count as horses?), and second, what makes qualitative identity (whether generic or specific) possible across numerical difference (e.g., how can red-horse-hood exist in both of these red horses at the same time when they are two horses rather than one?).
I understand Rand’s answer to the first question: red horses and brown horses possess different measurements of the same attribute, and we grasp the attribute by mentally omitting the measurements. But this can’t be her answer to the second question, since this solution, by helping itself to the notion of “same attribute,” presupposes that the second question has already been answered.
So what I’m wondering is: what is Rand’s answer to the second question? Does she even address the second question, or does she mistakenly think that all the philosophical fuss about universals has solely been about the first question? One reason for thinking she doesn’t quite see the second question is that when she first introduces the problem of universals (in Introduction to Objectivist Epistemology) she describes it this way:
When we refer to three persons as “men,” what do we designate by that term? The three persons are three individuals who differ in every particular respect and may not possess a single identical characteristic (not even their fingerprints). If you list all their particular characteristics, you will not find one representing “manness.” Where is the “manness” in men?
It’s clear from what Rand says here (e.g. the reference to fingerprints) that by “differ” and “identical” she means to signify qualitative difference and qualitative identity, not numerical difference and numerical identity. But in that case she’s missed half the question. Before we can start worrying about how it’s possible for two things to be qualitatively identical in the generic sense without being qualitatively identical in any specific sense, don’t we first need to justify the puzzling notion of qualitative identity per se?
2. In her 1964 article “Patents and Copyrights” (reprinted in Capitalism: The Unknown Ideal) , Rand offers inter alia the following argument:
As an objection to the patent laws, some people cite the fact that two inventors may work independently for years on the same invention, but one will beat the other to the patent office by an hour or a day and will acquire an exclusive monopoly, while the loser’s work will then be totally wasted. This type of objection is based on the error of equating the potential with the actual. The fact that a man might have been first, does not alter the fact that he wasn’t. Since the issue is one of commercial rights, the loser in a case of that kind has to accept the fact that in seeking to trade with others he must face the possibility of a competitor winning the race, which is true of all types of competition.
Here my question is this: does the patent office create the right, or merely record a pre-existing right? Because if the patent office creates the right, that seems to attributing to government a more sweeping authority than Rand ordinarily wishes to grant. But if instead the patent office records a pre-existing right, then that right, existing prior to certification by the state, cannot be lost by failing to receive such certification.
Nor is Rand’s analogy with commercial competition helpful. What I have on entering the market is not an unconditional right to sell my product, but only a right to try to sell it, or in other words, a right to sell it if I find a willing buyer. So if I am outcompeted by a rival seller who snaps up all my potential customers first, I haven’t lost any right. But if my rival beats me to the patent office, I do lose the right to try to find a willing buyer for my product (and the potential buyers likewise lose the right to try to buy from me). What justifies this?
After I wrote the above, I thought to look through my older writings on copyright to see whether I’d commented on Rand’s argument before. Turns out I did, and said basically the same thing:
Rand is suggesting that the competition to get to the patent office first is like any other kind of commercial competition. For example, suppose you and I are competing for the same job, and you happen to get hired simply because you got to the employer before I did. In that case, the fact that I might have gotten there first does not give me any rightful claim to the job. But that is because I have no right to the job in the first place. And once you get the job, your rightful claim to that job depends solely on the fact that your employer chose to hire you.
In the case of patents, however, the story is supposed to be different. The basis of an inventor’s claim to a patent on X is supposedly the fact that he has invented X. (Otherwise, why not offer patent rights over X to anyone who stumbles into the patent office, regardless of whether they’ve ever even heard of X?) Registering one’s invention with the patent office is supposed to record one’s right, not to create it. Hence it follows that the person who arrives at the patent office second has just as much right as the one who arrives first – and this is surely a reductio ad absurdum of the whole notion of patents.
Oh well, I guess there’s nothing wrong with having two different wordings of the same objection out there.
I’m not sure that I get what’s going on with your second question; its point may be eluding me. Let me try to restate it, with a different example, and see if I’m talking about the thing you’re talking about.
Here is a proton; it has a charge of +1. Here is an electron; it has a charge of -1. Those are different numerical values of the attribute of charge. But we say that it is the same attribute. How can we say that, when the positive charge is present in one particle, and the negative charge is present in a different particle?
Indeed, suppose we have a proton and another proton. How can we say that the positive charge of one particle is the same attribute as the positive charge of another particle, when those attributes exist in two different particles?
If that’s the question, I think the subatomic particles case may point at a possible answer. For what would it be for the attribute in the two particles not to be the same?
A positively charged proton and a negatively charged electron attract each other; two positively charged protons repel each other. If we envision a particle that is like a proton in all other respects, but instead of having a charge of +1, has a value of +1 of some different attribute—let’s call it, say, a pitch of +1—then the proton would not repel that particle, and conversely, it would not repel the proton. They would not interact either electrostatically or harmonicostatically (to coin an imaginary scientific adverb for an imaginary attribute). If we go further an imagine a whole cosmos filled with particles that each had its own unique set of attributes, not identifiable with the attributes of other particles, none of them would interact with the others, and they wouldn’t be identifiable as a cosmos.
Because they do interact, we can say that the trait that determines the nature of the interaction is the same trait in both of them.
This isn’t so simply applied to things other than subatomic particles. But we can say, for example, that two pieces of furniture both have the attribute of shape, even though their shapes are different, because we cannot put them into the same space in a room. We can say that two solutions both have the attribute of pH because they can neutralize each other. We can say that two stocks both have the attribute of price because we can sell one and buy the other.
Rand was a thoroughgoing realist, and didn’t deal in abstractly conceived metaphysical attributes. She focused on attributes by virtue of which things interact with other things, and with us. The attribute is the same if the things that have it are involved in the same network of causal relations.
Does that make any sense? Does it begin to address your puzzle in any way? Or am I just getting you entirely wrong?
“don’t we first need to justify the puzzling notion of qualitative identity per se?”
What would be your answer to that?
And how, if at all, does your answer to the first puzzle differ from Rand’s?
Is the notion really puzzling?
William H. Stoddard:
I’m not sure that I get what’s going on with your second question; its point may be eluding me.
Then we’re even, since I don’t understand your electron example either. If the electron and proton both have charge, but one has negative charge and the other has positive charge, that just looks like another case of generic identity across specific difference, which wasn’t the case I was puzzling about. How is it relevant to the case of qualitative identity across numerical difference?
Mark Twain and Samuel Clemens are one in number (because they’re just one person, not two). Two particles may be one in some quality (say charge) but they’re not one in number because they’re two particles, not one. The problem is: if, as Rand says, everything that exists is particular, then what does “same” mean in the claim “these two particles have the same property”? It doesn’t seem to mean what “same” means in “Mark Twain and Samuel Clemens are the same person” — that is, there’s presumably no numerically identical thingy that’s present in both particles simultaneously. Otherwise, as Plato puts it, “one and the same thing will exist as a whole at the same time in many separate individuals, and will therefore be in a state of separation from itself.”
What would be your answer to that?
For my own take see this post.
And how, if at all, does your answer to the first puzzle differ from Rand’s?
I have no major disagreement with Rand on the first point; see my abstraction article.
Is the notion really puzzling?
I think it’s puzzling only if one thinks about it in the wrong way. As traditionally asked, the question is unanswerable, because it presupposes that impositionism and reflectionism are the only possible positions with regard to universals.
So Rodderick, did you ever get around to reading Scott Ryan’s critique? The audience is waiting with breathless anticipation for your analysis.
“As an objection to the patent laws, some people cite the fact that two inventors may work independently for years on the same invention, but one will beat the other to the patent office by an hour or a day and will acquire an exclusive monopoly, while the loser’s work will then be totally wasted.”
I know very little about patent law, but I believe if you are sued for infringing on a patent, you may assert that you developed the invention separately.
Thanks Roderick. I read both the post and an unpublished draft of the QJAE article as well as your book draft on Wittgenstein and the related RAE article a couple of years ago, but it is good to refresh my memory and tie them into the issue of universals. Curse my poor memory.
How is it relevant to the case of qualitative identity across numerical difference?
Well, take two protons, which both have a charge of +1. They are two different particles. So what does it mean to say they have the same charge?
Well, it means that they repel each other with a certain force. And it means that if you bring an electron near them, they both attract it with the same force. And it means that if you set them in motion, they both generate magnetic fields. And so on. Whereas if you do this with a proton and a neutron (charge 0), the neutron won’t repel protons, or attract electrons, or generate magnetic fields. “The same” means, roughly, causally indistinguishable from each other.
But isn’t that circular? You say that what makes two properties the “same” is that they’re causally indistinguishable. I was once tempted by that view. But to say they’re causally indistinguishable isn’t to say that they produce the numerically same effect; it’s to say that they produce the same kind of effects, i.e. effects that are qualitatively the same along some dimension. But “same kind” is precisely the concept that was supposed to be explained.
For example, I could say that these two red splotches have qualitatively the “same color” insofar as they produce the “same sensation” in me. But by same sensation I don’t mean numerically the same; after all, they produce two sensations, not one. I must instead mean that they produce two sensations that share a common property. But the original puzzle was precisely to explain how one property can exist in two things simultaneously if everything that exists is particular. Moving the explanandum from the cause to the effect — e.g. from the splotches to the sensations — doesn’t seem to help.
I don’t know. It seems to me that a certain concept of sameness arises naturally out of the observation of physical reality. Whereas it seems that the concept of sameness you’re asking for is some sort of Platonic ideal of sameness that could never be identified in any two things. That way lies the “third man” argument. . . .
In which case, the Randian response might be to say that you can’t prove it, because you must accept it as true to form any concept whatever, or to prove anything whatever. That is, it may be another axiom, though one that Rand didn’t explicitly name. (It wouldn’t be the first one; the reliability of memory appears to be an axiom, in that conceptual knowledge is impossible without it, but Rand never names it as such that I’ve been able to find.)
I don’t think I’m asking for a Platonic ideal. I’m asking about the plain ordinary sameness we find all over the place. I agree that it can’t coherently be denied. But as I mentioned, I don’t think the sameness I’m talking about is philosophically problematic, because I think it’s what Wittgenstein would call a showable, not a sayable. My question is how Rand can reconcile it with her own system, which doesn’t seem to include unsayable showables, and which holds that everything that exists is particular?
“which doesn’t seem to include unsayable showables,”
Perhaps this is way off base, but… ostensive definitions?
“With certain significant exceptions, every concept can be defined and communicated in terms of other concepts. The exceptions are concepts referring to sensations, and metaphysical axioms.” (ITOE, p. 52)