Notes on Kripke, "Naming and Necessity," Lecture 2

 1. Quick catch-up

Kripke begins the second lecture by summarising his main points from the first, claiming that the intuitions supporting the theories of transworld identification and counterpart theory are bizarre. 

1. To every name or designating expression 'X', there corresponds a cluster of properties, namely the family of properties 𝞅 such that A believes '𝞅X'. 

2. One of the properties, or some conjointly, are believed by A to pick out some individual uniquely.

3. If most, or a weighted most, of the 𝞅's are satisfied by one unique object y, then y is the referent of 'X'. 

4. If the vote yields no unique object, 'x' does not refer. 

5. The statement, "If X exists, then X has most of the 𝞅's" is known a priori by the speaker.

6. The statement, "If X exists, then X has most of the 𝞅's" expresses a necessary truth (in the idiolect of the speaker).

He says that, for him, Aristotle’s most important qualities were his philosophical works, and Hitler’s were his murderous political role, and both of these are contingent, not necessary, facts about the men. Thus, he says, we must cross (6) off the list of acceptable tenets. Yet Kripke is not being careful, because he is discounting the biological properties that are presupposed by the properties he claims are most important. Certainly if a property depends (asymmetrically) on another for its existence, the dependent property is less important than the one on which it depends. Therefore, the properties he exalts in these men cannot be their most important properties. He is talking about the properties that are most significant to him personally, but that does not make them the properties that are most important to fixing the referent.

He then claims that tenets (2)-(5) are only all true in rare cases, which he says are usually “initial baptisms.” He claims (2)-(4) have a large class of counterexamples, and that (5) is usually false. He says the truths of (3) and (4) are empirical accidents and “hardly" count as a priori knowledge. This is all very bizarre, since none of these conclusions are valid deductions based on the reason from the first lecture. I should note, however, that I suggested a change to (3) to accommodate a potential difficulty that Lewis anticipated in his counterpart theory:

(3*) If most, or a weighted most, of the 𝞅's are satisfied by one unique object y, then y is the referent of ‘X’, unless y uniquely satisfies most (or a weighted most) of the conditions for a referent of 'Z', where X ≠ Z.

2. A picture of naming

Kripke says that the counterfactual theory and the cluster concept theory should be understood as providing a picture of naming. He says the picture is no doubt correct for a limited set of cases:  first, in cases where we attach a name to an object already identified by description (as in the case of “the man who denounced Catiline”) as well as in cases where an object is identified ostensively (either directly by pointing, or indirectly by physical evidence). But these are unusual cases, he claims. He says, the theses do not hold for many, or even most, actual cases of naming.

Kripke proceeds to knock down a straw man version of (2). He supposes that a person who takes “Feynman" to refer to a famous physicist, without knowing anything else about him, would (according to (2)) therefore assume that “a famous physicist” picks out one person uniquely. As if only knowing only that Feynman was a famous physicist would lead one to believe that Feynman were the only possible famous physicist! This is not what (2) means. It explicitly indicates the belief that a set of properties uniquely picks out a person. Nobody in their right mind would believe that “a famous physicists” uniquely picks out anybody.

Now we get to a discussion of Godel and Schmidt (who, hypothetically, is the real person responsible for Godel’s Incompleteness Theorem), which is similar to the example of Peano/Dedekind. But all these examples show is that we should not take historically significant achievements as carrying very much weight when we consider the reference-fixing properties of a name. Kripke assumes that historical achievements should be sufficient, even though he previously acknowledged that weighting is critical. He has not made one step toward offering a reason for weighing one property over another.

3. My own thoughts on weighting:

In my previous notes, I indicated that common-sense weighting rules are advisable. We could state some as follows:

(W1) If two distinct properties share an asymmetrical relation of dependence, then the more dependent property is weighted less than the more fundamental property.

(W2) If a property is entirely dependent on another for its existence, then the dependent property is weighted less than the one on which it depends.

(W3) If a property is such that multiple properties depend on it, then that property is weighted more highly than the sum of the weights of the dependent properties.

Since historical achievements and other cultural properties depend entirely on biological properties, then (following W3) the combined weights of those achievements and cultural properties are less than the weight of the more fundamental biological properties.

We might be tempted to say that biological properties rely entirely on physical properties, but this is a confused way of talking. Biological properties are physical properties. We might then try to clarify as follows: Biological properties are dependent on chemical properties, which are in turn dependent on more fundamental physical properties. However, it is not evident that we must regard all of these properties as distinct from biological properties. The fact that chemistry and physics can, in an abstract sense, explain biology is not sufficient to conclude that biological properties depend on chemical or other more fundamental properties. We might say that biological properties are manifestations of such properties, but that is not the same thing. So biological properties seem to be the most fundamental—when it comes to fixing the reference of proper names of people, at least. And all other properties rely in whole or in part on biological properties. Therefore, these should comprise the majority of the weighting.

We might still suppose there is no difference between the relationship between biology and chemistry and the relationship between historical achievements/culture and biology. Aren’t Aristotle’s achievements manifestations of his biological properties? In a sense, yes, but not in the same way that biology is a manifestation of chemistry and physics. Aristotle’s achievements are dependent on biological properties that describe Aristotle apart from his historical achievements. This is not the case with biology and culture. We would not say that biology is dependent on chemical and other physical properties that also describe Aristotle apart from his biological properties. Rather, those chemical and physical properties are Aristotle's biological properties. 

So there is a common-sense way of understanding these weighting principles that appears to stand up to scrutiny, and which reveal severe flaws in Kripke’s thinking.

4. On Strawson and chains of reference:

Kripke says Strawson is trying to pass the buck by claiming one’s reference may derive from another’s (p. 90). Kripke claims that this leaves reference in a state of uncertainty, and therefore will not do. But he does not bother to consider how this actually happens. As I noted in my comments on his first lecture, a student makes assumptions about referring terms. If asked who Aristotle was, they will defer to the experts in the field, of whom they take their teacher to be a competent representative. In any case, he is misrepresenting Strawson, as we shall soon see.

On the next page (p. 91), Kripke offers an alternative to this view: when babies are named, the name travels from person to person, “from link to link as if by a chain.” Kripke allows that a person may come to learn of a person and not remember how exactly they learned of them. But they will remember the name, and apparently that is enough to retain a sense of reference. (This is counterintuitive, of course. If all I remember is a name, I will not have any sense of reference at all. To use the name, I need to have some sense of what would count as an appropriate use of that name. But this is precisely what Kripke denies . . .)

A page later, Kripke says this is different from Strawson’s account, but he only gives us a straw man version of Strawson's views. He claims Strawson requires that one who has learned a name this way must be able to say from where their description was obtained. This is obviously not required of a descriptive theory of names in general, and it is not what Strawson says in the footnote that Kripke quotes. 

Strawson says (it is in a footnote on page 182 of Individuals, Routledge, 1959) that one may include a reference to a different person as a source of a referring term (what Strawson is specifically referring to are “identifying descriptions”), and that in so doing, one does, in fact, pass the buck. That does not mean that one’s ability to refer depends on their ability to remember where and when they learned a referring term! Strawson says exactly the opposite in the very same footnote (which Kripke conveniently excludes from his lecture): “It is not required that people should be very ready articulators of what they know." 

Kripke claims to be establishing something new and intuitively preferable to descriptivist views (including the cluster concept view) by appealing to a chain of reference (though on p. 93 he tries to walk back the extent to which he can be said to be offering a theory). The point Strawson was actually making in that footnote is that, if one wants to offer an identifying description, one must be able to actually describe something. Kripke does not suppose anything about what one must know to properly use a referring term. He says that's a matter for a fully developed theory, and he has no interest in that. He actually says he's too lazy to do it. I'm sure that line was delivered with a sense of humour in the lecture, but it is still rather frustrating. He doesn't seem to realise how much of a hole he is digging for himself.

5. Kripke’s theory of naming

Resisting pressure (real or imagined) to provide necessary and sufficient conditions for his picture of naming, he offers a rough statement of his theory: 

(K1) Initial baptism: a name is fixed by ostension or description;

(K2) The name is passed on: a receiver of the communication intends to refer to the same individual as the speaker.

Kripke observes (p. 97) that he is referring to reference to explain the transmission of names. Kripke distinguishes this from the descriptionist view which he says (footnote 44, p. 97) requires "that each speaker essentially uses the description he gives in an initial act of naming to determine his reference.” Is Kripke saying that a descriptionist cannot account for learning names? That a descriptionist believes we actively describe things by referring to them?  To use a description in an act of naming is to attribute a property to that which is named. Descriptionists can distinguish between referring and attributing properties. What they claim is that the act of referring implicates some properties—that the name, as used by the speaker, picks out an individual because of some known properties (whether or not the speaker is ready to explicate them).

Kripke would apparently have it that once you learn a name, you can use it to refer to whoever the person who taught it to you was referring to—regardless of what you remember about it. If Kripke wants his chain of references to mean something—to allow somebody to actually refer to somebody—then they must be said to know something about that somebody, even if (as Strawson allows) they cannot readily articulate what they know. If they use the name but cannot say anything at all, or demonstrate anything at all, about how it should be used, then we would not say they are using the name correctly. They are looking for a correct use, and maybe they will remember one.

Imagine a person learns of Cicero and also learns of Aristotle, then forgets who was who. They refer to Cicero as "the student of Plato who went on to write notable works of philosophy and become a teacher of Alexander the Great." We would not say that they are talking about Cicero at all, even though Kripke apparently thinks they must be. We would say that they are talking about Aristotle and mistakenly referring to him as Cicero. 

We would say, “That was Aristotle, not Cicero!” 

We would not say, “No, that is false. You don’t know anything about Cicero!” 

This is because we take them to be referring to a person by virtue of the descriptions they are using, and which we recognise as describing Aristotle, not Cicero. By detaching names from descriptions, Kripke cannot account for this. He therefore has not shown any superiority to his “theory”—if it really is a theory of naming, and not simply an observation that people can learn names from other people! (As if that was in doubt!) All he has done is rejected respectable attitudes toward descriptions by misrepresentation.

6. On identity statements

Kripke abruptly changes the subject on page 97. He talks about identity claims (Hesperus is Phosphorus, Cicero is Tully, Heat is the motion of molecules, . . .) which, he says, are commonly thought to be contingent facts, but which are actually necessary facts (he does not distinguish between physical and metaphysical necessity here, saying that they might be the same for some sorts of facts). He rejects the identity thesis, according to which conscious experiences are neuro-physiological states. He says this is a contingent sort of relation, and therefore not like the fact that heat is the motion of molecule. Why? He does not say today.

 

On the names “Hesperus" and “Phosphorus" being used to refer to the evening star and the morning star, respectively, Kripke says they are rigid designators—they refer to Venus in all possible worlds. I want to see if this holds.

Let us imagine that somebody uses “Hesperus" to designate the experience of seeing a celestial body in that position in the morning, and they use “Phosphorus" to designate the experience of seeing a celestial body in that position in the evening. These are experientially different, and so the different names have different senses. They have different applications.

Now, somebody could come along and say that the same celestial body is responsible for both experiences, but why should this matter? Our imaginary person can have a particular worldview such that every celestial body is of the same substance. In their view, every celestial body is a manifestation of the same unitary being. Therefore, our person already believed that Hesperus and Phosphorus were both, in some sense, caused by the same celestial body: There is only one celestial body. They still distinguish between Hesperus and Phosphorus because they are reliably distinct experiences: one in the morning, one in the evening, with easily distinguishable appearances. 

Appeals to science are simply irrelevant. Our person may accept that, from a certain point of view, both Hesperus and Phosphorus are names of a planet called Venus, and this can be studied in all sorts of ways. They will not doubt any of that. They simply mean by “Hesperus” a certain experience in the morning, and by “Phosphorus" a certain experience in the evening. They can pick it out very easily, without any confusion. And nobody has any trouble understanding them.

Kripke says that these words refer to the planet Venus in every possible world, but this is to assume that a person is referring to one particular scientifically identifiable aspect of an experience, and not the experience as a whole. The person who names the experience, and not the planet, will not see any scientific analysis as salient, and so they will not care that it happens to be a planet called Venus over there. They will not see “Hesperus" as picking out a planet at all!

7. Kripke’s counterintuitive and problematic position

Kripke wants us to think about a different case than the one I have described. In his scenario (p. 104), we use “Hesperus” and “Phosphorus" to specifically identify the same celestial body. In this case—in the case in which we use two names to identify the same thing—then we can regard those names as identifying the same thing in every possible world.

Kripke refers to two “qualitatively identical epistemic situations”—meaning, I take it, that they are experientially the same for the individual, producing the same knowledge: one in which “Hesperus" and “Phosphorus" refer to two different celestial bodies and one in which they refer to the same. Thus, we need an empirical investigation to determine which case is true for us. So there is a sense, he says, in which either possibility could have turned out true. That is, both were possible with respect to our knowledge/experience. But then he immediately says the opposite, that it is not the case that it could have turned out either way—and I suppose in this sense he means in the sense of objective reality, because the two names refer to the same planet. But this is only to say that what is possible with respect to our knowledge/experience is not the same as what is actually the case.

Kripke makes a questionable claim here. He says that, because we are using the two names to unknowingly refer to the same object, then a counterfactual world in which they refer to different objects would not match our current usage. The implication is that our usage is not a matter of what we know or intend, but a matter of what we actually pick out in the world. This seems problematic. (In a footnote, he says that this is developed more in the third lecture.) 

The point here is related to the two different senses of “sense" that he is interested in. In one sense, there is the epistemic situation according to which we can stipulate possible worlds. In this situation, we can imagine that Hesperus might not be Phosphorus. I would say this is because the manner in which we pick out Hesperus is not the same as the manner in which we pick out Phosphorus. That is a proper Fregean sense, I think. But then there is the sense of fixing a referent, and Kripke wants to say that this is different. Apparently, the way it is different is that it is not dependent on our epistemic situation at all. The referent is what it is regardless of what we think about it. This is mind-independent reference. In this sense, either Hesperus is Phosphorus or it is not, but we cannot say both are possible. Only one is the case, and the other is a mistake.

Next, he says that the following is true:

(K3) If Hesperus and Phosphorus are one and the same, then in no other possible world can they be different.

The meaning of this is apparently that 

(K3*) If the object H identified by the name “Hesperus" is the same as the object P identified by the name “Phosphorus,” then H = P in all possible worlds (or, to put it another way, H = P necessarily).

As it stands, I don’t see a problem there. But he wants to say something about the names:

(K4) If Hesperus and Phosphorus are one and the same, then we have to use the names “Hesperus" and “Phosphorus" to refer to that same body in all possible worlds.

Now why would he say that? Clearly we might not know that they are one and the same, so why should our use of the names depend on them being the same? Indeed, K4 leads us to a bizarre situation. Assuming we do not know that both refer to the planet Venus, we can suppose that Hesperus is not Phosphorus without contradiction. Yet, if K4 is true, there is a contradiction. Supposing that Hesperus is not Phosphorus is just to suppose that Hesperus is not Hesperus. That is surely not what we are stipulating. This makes K4 both counterintuitive and problematic. (This is precisely the problem Frege solved by introducing the sense/reference distinction.)

His reasoning seems backwards. After going from K3 to K4, he then suggests that K3 actually depends on K4. He says, "If, in fact, they are the same body, then in any other possible world we have to use them as a name of that object. And so in any other possible world it will be true that Hesperus is Phosphorus." It does look like he is saying K3 follows from K4, which is bizarre. K3 is independently true, and K4 seems independently false.

Kripke ends this lecture by briefly discussing the contingency of Hesperus and Phosphorus being used to refer to the same planet. It’s possible that Venus might not have been visible in the morning, so that the name “Phosphorus" might have picked out something else. So, he says, it is not necessary that the words refer as they do; but since they do refer as they do, they necessarily refer to the same object.

The question is, how can Kripke say this? It’s like he is trying to speak in two different idioms at the same time. In one idiom, Hesperus and Phosphorous are only contingently the same; in the other idiom, they are necessarily the same body. And since he is using the latter to enforce restrictions on how the names “Hesperus" and “Phosphorous" can be used, we simply cannot say that they are only contingently the same. So we cannot speak in the first idiom at all. 

To see why this is a problem, consider what Kripke says at the end of the second lecture:

Of course, it is only a contingent truth (not true in every other possible world) that the star seen over there in the evening is the star seen over there in the morning, because there are possible worlds in which Phosphorus was not visible in the morning. 

He is apparently being careful to use demonstrative expressions (“the star seen over there”) instead of proper names in the first clause of this sentence. However, as he has acknowledged before, such expressions are still used to fix a referent. In this case, the two instances of “the star seen over there” fix the same object, which is Venus. Therefore, by Kripke’s rules, we cannot say that it is a contingent truth that they are the same body. The fact is that they are the same body, and so “the star seen over there in the evening” is necessarily the same body as “the star seen over there in the morning.” Why should he be allowed to use these two phrases to pick out a contingent truth while reserving “Hesperus" and “Phosphorous" to pick out a necessary truth?

He needs a theory here, but it looks like he might have undermined any hope of finding one.