THE ROLE OF NAMES IN SCIENCE
The function of names, at least, appears straightforward: it is to refer to an individual. When we assert that 'Max Planck was the first to quantize energy', we are using the name 'Max Planck' to refer to a specific individual. The name labels the individual. Likewise, when we write the expression |a 1 >, the label 1 is being used to refer to an individual, or so it would appear. A number of well-known philosophical issues then arise concerning reference, meaning, and the 'mechanism' by which names refer. 1
Let's begin with an apparently straightforward view of the relationship between proper names and their bearers, according to which names simply stand for, 'tag' or 'label' objects. They have no meaning other than standing for these objects. Hence, a name does nothing more than denote its bearer and its meaning is entirely given by its reference. As Mill put it, proper names have denotation but not 'connotation', in the sense of connoting a set of specified properties. By comparison, a kind term such as 'horse' has both denotation and connotation: it denotes the kind, and connotes a set of properties which are specified in the definition of 'horse'. A proper name, on the other hand, simply and only denotes its bearer. As Wittgenstein subsequently wrote, "[a] name means an object. The object is its meaning". 2 For reasons which will shortly become clear, this can be called the 'Non-Descriptivist' view of proper names.
How do names refer on this view? The answer appears to be straightforward: they are like labels and are attached to or assigned to their bearers, just as labels are attached to an object; or better, just as a person's name is assigned to them at birth. There is then the further issue of how we come to know who or what the bearer of the name is and again, thinking, for example, of how we would come to know who 'Max Planck' refers to, we can come up with some fairly obvious answers. Already this is to fill in some of the details of the 'mechanism' of reference and we will return to these issues shortly, but first we must consider some of the well-known problems this view must face.
Let us consider again the example we mentioned in Chapter 1. 'The Morning Star is identical to the Evening Star'. This is an example of an identity statement which conveys information. How can that be, when the two names are referring to the same object? From the Non-Descriptivist perspective, such statements are trivial. Or consider the statement 'Victor Jakob could not accept the new quantum physics', 3 which appears meaningful, but has to be regarded as meaningless on this view. And the assertion 'Victor Jakob does not exist', which appears to be not only meaningful but, intuitively, true, must be dismissed as absurd, since to be meaningful, it must presuppose the existence of that whose existence it denies! Of course, an adherent to the Non-Descriptivist position could always maintain that in each case it is our intuitions that are wrong-that indeed the proposition expressed by 'The Morning Star is identical to the Evening Star' is trivial but that nevertheless information is still communicated by this assertion, that although the statement 'Victor Jakob could not accept the new quantum physics' does not express a proposition, nevertheless some proposition is communicated by its assertion, and so on. However, this requires some account of how information can be communicated by trivial or meaningless propositions.
Given these difficulties, one might turn to the alternative, 'Descriptivist' view, which holds that names refer by virtue of the descriptive content associated with the name, where this content is typically specified by means of a definite description satisfied by the object answering to that name. For this 'mechanism' to work the description (or least some aspect of it) has to uniquely pick out the object being referred to. Of course, different descriptions may be associated with a name, both between individuals using the name and-more significantly perhaps-over time. Thus a historian of science may associate the description 'the person who first regarded energy as quantized' with the name 'Max Planck', while a student has the vaguer association 'some German scientist'. Obviously if there is sufficient overlap between these descriptions we would typically be happy to accept that these individuals are ref 242x2321c erring to the same person. In the case of shifts in descriptive content over time, a well-known problem arises for those who would like to be realists about physical objects: if the shift in descriptive content is sufficiently radical that there is no common element between descriptions before and after the shift-as in the case of scientific revolutions, perhaps-then it might be asked, on what grounds can it be claimed that the different descriptions are referring to the same object? Such concerns are precisely the motivation for realist-inspired developments in theories of reference, which we shall touch on shortly.
Before we do, let's just consider some of the virtues of this view. First of all, it offers an account of the meaning of proper names: in place of the bearer of the
end p.200
name constituting its meaning, the Descriptivist maintains that the meaning is given by the descriptive content which fixes the reference. Thus we recall from Chapter 1 Frege's claim that the name-referent relation involves a third element-the sense of the name, which provides the meaning: "[a] proper name expresses its sense, stands for or designates its reference". 4 He argued that names essentially have a sense, but only contingently have a reference-they refer if and only if there is an object which satisfies their sense. The sense of the name can then be explicitly formulated by means of a definite description-a list of properties-and thus proper names come to be seen as equivalent in meaning to, or can be viewed as no more than, 'disguised' definite descriptions. One way of summarizing the difference between the two views we have outlined is to say that on the Non-Descriptivist theory, naming is taken to be prior to describing, whereas on the Descriptivist view, describing is prior to naming (for a name only names by describing the object it names). As Searle also noted, they point in different directions, metaphysically speaking:
These two views are paths leading to divergent and hoary metaphysical systems. The first leads to ultimate objects of reference, the substances of the scholastics and the Gegenstände of the Tractatus. The second leads to the identity of indiscernibles, and variables of quantification as the only referential terms in the language. 5
Returning to the virtues of Descriptivism, it can apparently handle the problems faced by the Non-Descriptivist. So, if we consider again, 'The Morning Star is identical to the Evening Star', the Descriptivist theory precisely accounts for how such a statement can convey information even though the two names refer to the same object by positing that the senses, or meanings, of the names are different. Furthermore, as far as the statements 'Victor Jakob could not accept the new quantum physics' and 'Victor Jakob did not exist' are concerned, the Descriptivist view demonstrates how one can meaningfully use a name without supposing that there exists the entity so named. As Quine put it: ". where descriptions are concerned there is no longer any difficulty in affirming or denying being". 6 In this manner, a statement of existence or non-existence is not seen to contain any expression naming the entity whose existence may be in question. Thus, whether such a statement means anything
end p.201
or not does not presuppose that such an entity actually exists:
We need no longer labor under the delusion that the meaningfulness of a statement containing a singular term presupposes an entity named by the term. A singular term need not name to be significant. 7
Nevertheless, Descriptivism also faces difficulties. We've already noted that different people may attribute a different sense to the same proper name, as Frege himself realized. 8 Furthermore, there may be more than one possible description of the object which is available and it may be unclear to which of these the name should be associated. In particular, if contingent properties are included in such a description, then we run into the problem that any true statement about the object that used the name as subject would be analytic and any false statement would be self-contradictory. If, for example, the name 'Max Planck' is taken to mean 'the person who first regarded energy as quantized', then the statement 'Max Planck was the person who first regarded energy as quantized' becomes a tautology on this view. This is surely counter-intuitive, as the statement appears to express something which, with a little historical investigation, could be shown to be false. 9 And finally, the meaning of a name and perhaps the identity of the object would have to be regarded as changing every time there was a change in the properties of the object and the same name would have different meanings for different uses of the name.
Hence we have a dilemma: if we consider the role proper names play in language, then the Non-Descriptivist theory, in asserting that such names essentially refer but have no sense, seems to have it right; but if we think about the 'mechanism' of reference, then Descriptivism, in claiming that such names refer only if one and only one object satisfies the relevant description, seems to be correct. Again, as Searle puts it:
The subject-predicate structure of the language suggests that the [Non-Descriptivist theory] must be right, but the way we use and teach the use of proper names suggests that it cannot be right: a philosophical problem. 10
Searle's solution is to cash out the descriptive content associated with a name in terms of a sufficient but nevertheless unspecified number of 'standard
end p.202
identifying statements', where the use of the name to refer presupposes the existence of an object of which a sufficient but unspecified number of 'standard identifying statements' are true and to be that object is to possess a sufficient but unspecified number of the relevant properties mentioned in the 'standard identifying statements'. Of course, one might want to insist that certain of these statements carry more weight than others. However, the important point is that the lack of specificity in the number of 'standard identifying statements' is a function of the role played by proper names and to fully pin down these statements would be to reduce names to nothing more than descriptions and undermine this role. To ask for the full set of criteria for applying a name is to ask for the full set of identity criteria of the object. If such a full set of criteria were to be given, we would have a complete description of the object, and we could then only refer to it by describing it. But pragmatically, names don't function as descriptions like this but rather act as 'pegs on which to hang descriptions'.
This, Searle claims, resolves the problem: names do not have sense or meaning in the sense (ha!) that they are used to describe objects but nevertheless they are logically connected with the relevant characteristics of the objects to which they refer, in a 'loose sort of way'. Take the statement 'The Morning Star is identical to the Evening Star' one more time; then, Searle insists, this statement might well be analytic and uninformative for some people, since the same descriptive content may be associated with each name, but, of course, in many cases this content will be different and then the statement will be not only synthetic in these cases, but "might even advance a historical discovery of the first importance". 11 Or consider again the statement 'Viktor Jacob does not exist'; this is now to be understood as asserting that a sufficient number of descriptive statements associated with referring uses of the name 'Viktor Jacob' are false. Hence we can understand how the statement tells us more than simply that the name was never used to refer to any object; that is, we can understand how this statement is meaningful. Nevertheless, we can accommodate the points that different people may attribute a different sense to the same proper name and more than one possible description of the object may be available, without it being clear which should be associated with the name.
However, Searle's view also has unpalatable consequences. Let us take the name 'Albert Einstein' and the statement, attributed to Einstein, 'If only I had known, I should have become a watchmaker'. Is it possible for Einstein to have been a watchmaker? Intuitively, this certainly seems a possibility, one whose strength might be increased by the discovery that Einstein was handy
end p.203
with tools, took watchmaking classes, etc. But on Searle's view, if Einstein had never been a physicist, he would not have possessed a crucial property mentioned in one of the 'standard identifying statements' which should surely carry more weight than perhaps any other; in other words, he would not have been Einstein. Searle himself bites the bullet here and insists that ". it is a necessary fact that [Einstein] has the logical sum inclusive disjunction, of properties commonly attributed to him .". 12 However, if we grant legitimacy to our modal intuitions on this and accept that it is possible for Einstein not to have been a physicist, then some account is required which allows us to refer to this person who was not a physicist but a watchmaker as 'Einstein'. 13
Kripke has attempted to provide just such an account via the notion of a 'rigid designator' which designates the same object in every possible world, whereas a non-rigid or 'flaccid' designator, such as a description, does not. 14 Proper names are rigid because they satisfy this 'intuitive test'. As Kripke put it: ". although someone other than the U.S. President in 1970 might have been the U.S. President in 1970 (e.g. Humphrey might have) no one other than Nixon might have been Nixon" 15 and furthermore, ". although the man (Nixon) might not have been the President, it is not the case that he might not have been Nixon (though he might not have been called 'Nixon')". 16
The central idea, then, is the following: we fix the reference of a proper name by an 'act of baptism', in which the object is named by ostension, or the reference is established by a description. The latter does not yield the meaning of the name, but simply fixes the reference via certain features of the object which the object has contingently, in this, the actual world. When this name is passed along the chain of communication, the person who receives it must then intend to use it with the same reference as the person from whom he or she heard it. Whether established by ostension or description, the name then rigidly designates the object which possesses this property in this world, in all possible worlds. Even if, in such possible worlds, the object no longer possesses the property used to fix the reference in the actual world, the proper name still rigidly designates that object. In other words, once its reference is fixed in the actual world, the designation of the proper names is unaffected by possible changes in the actual world which we can imagine. 17 Underlying this view, of course, is a particular theory of the nature of possible worlds which holds that such worlds are not 'out there' as it were, to be viewed through some kind of metaphysical telescope, but are effectively constructed by us. 18 Thus, with the referent of 'Nixon' fixed in this world by, say, the property of being U.S. President in 1970, we construct a world in which Nixon does not possess that property but is a second-hand car salesman, say. Likewise, we may pick out the referent of 'Einstein' in the actual world as the physicist who proposed the theory of Special Relativity. In a possible world in which someone else achieved this result-Lorentz, say, or Poincaré-we would not refer to that someone else as 'Einstein'. If Einstein had never become a physicist, but had obtained gainful employment as a watchmaker, he would not have possessed this property which we use in the actual world to fix the reference of his name. This is precisely because the name 'Einstein' rigidly designates a certain person. Even though the property we have used to fix the reference is an accidental one, we still use the name to designate the person in all possible worlds.
Kripke's account is a version of the 'Causal' theory of proper names, so called because the original reference is recovered through a causal chain of communication by means of which someone using the name can be understood as intending to refer to that individual which was originally baptized with the name. According to this account, identity statements, for example, are, indeed, necessarily true, since the reference is the same, but they are not known a priori to be true, since the descriptions are different. 19 Thus we do not know a priori that the Morning Star is identical to the Evening Star, because the evidence by which we fix the referents of the names allows for the possibility that these referents might be different. The necessary truth of such identity statements ultimately expresses a relation between the object and itself-that is, self-identity-rather than a relation between names. Existential statements can also be dealt with: 'Victor Jakob does not exist' is not absurd, nor is 'Victor Jakob could not accept the new quantum physics' meaningless, since the relevant fiction or story in which the name appears substitutes for a 'baptism' and thereby supplies the meaning. 20
end p.205
This view was proposed, at least in part, as a response to the problem of theory change noted above. It shifts focus from the apparently dramatic switch in theoretical descriptions across revolutionary divides, to the causal chains of communication tied to some original baptism event. The idea, then, is that even though, say, the theory of the electron has changed dramatically, we can still claim to refer to the electron through such changes as long as these causal chains are maintained. As it stands, however, this view still faces a problem of reference change. Consider the following example (from Hacking): in 1935 Yukawa proposed a theory of the nucleus which predicted the existence of a particle responsible for nuclear forces. This particle was named the 'meson'. In 1936 Anderson and Neddermeyer discovered a particle through observations of cosmic rays, which was identified with the meson. However, subsequent observations of the behaviour of this particle led to the conclusion that it was not Yukawa's meson at all and so it was re-dubbed the 'muon' (it was originally termed a 'mu-meson'). It was subsequently realized that the particle predicted by Yukawa was the 'pion', discovered in 1947.
What we have here is a case where a name is originally
referred to one kind of particle, and is subsequently attached to another. When
Nevertheless, this Causal theory remains open to the objection that it needs to be supplemented with a theory of meaning which can account for the fact that names do appear to have cognitive content. Thus, for example, there is still something rather unsatisfactory with its treatment of identity statements since the difference between 'The Morning Star is identical to the Evening Star' and 'The Morning Star is identical to the Morning Star' remains opaque. Yet this is the crucial difference that any account in which proper names themselves have reference but no meaning, seems incapable of explaining. Likewise, it is unclear how, precisely, the relevant fiction or story supplies meaning to
end p.206
existential statements when the names themselves have no such meaning. 22 An obvious way of spelling this out would be in terms of the story providing an appropriate context but now there is the danger of this collapsing into a form of Descriptivism: of course, if we substitute 'the fictional protagonist of McCormach's novel who finds himself unable to cope with the changes in physics at the beginning of the twentieth century' for 'Victor Jakob', then it is clear that 'Victor Jakob does not exist' is only contingently true but the victory has been won at the price of conceding the explanation to Descriptivism. Furthermore, in order to fix the reference of a name, the dubber must at least know what kind of object it is. But that involves the satisfaction of a certain predicate and that means conceiving of the object in terms of a certain mode of classification, or simply, a description. 23
These objections suggest that some descriptive element must be introduced into the account and hence various 'hybrid' theories have been proposed. Without going into too much detail, the general idea is that the reference of the name is now seen as the main causal source of the description which the speaker associates with the name. 24 Consider, for example, the case of a future (and possibly alien) historian of science who discovers a fragment of text in which the theory of Special Relativity is presented and which gives the name 'French' as the author of the text. The history of science community then uses the name 'French' to refer to the person who proposed the theory of Special Relativity, since their intention is not to introduce any novel use of the name and they believe that French did indeed propose the theory. However, it might be claimed that intuitively the name refers to the physicist formerly known as Einstein and not to the author of a well-known undergraduate textbook on Special Relativity. Hybrid theories handle this by insisting that it is the physicist who constitutes the 'dominant causal origin' of the description associated with the name, namely 'the person who proposed the theory of Special Relativity' and hence the name actually refers to that physicist.
It will be useful for our purposes to focus on Gracia's version of the hybrid approach, which explicitly and neatly incorporates the distinction between individuality and distinguishability. 25 According to this view, there are two aspects to the process of naming or 'tagging' an individual: the first concerns our epistemic access by means of which we become aware of the individual,
end p.207
the second is to do with the act of denoting, or giving the individual a name. The former involves description, whereas the latter ". has no descriptive mediation". 26 According to this approach, the primary function of proper names is to refer and the symbols employed refer not in virtue of any descriptive meaning they have but simply because they stand for the individuals concerned, where this individuality is understood in terms other than those of some set of properties, of course. 27 When this approach is applied to proper names,
. we can see not only that proper names need not be descriptive in any sense in order to have referents, as descriptivists believed, but also that it makes sense to consider them primarily nondescriptive and therefore as devoid of meaning and sense. 28
If we return to the issue of explaining how names refer, then from Gracia's perspective the concern here is simply misplaced. To seek a mechanism of reference is to misunderstand the primary function of names. Since it was this concern, this desire for an appropriate mechanism, which motivated the introduction of meanings via some form of descriptive element, there is now no need to hold that any such meaning is required in addition to reference.
When it comes to the issue of how names are established, however, we need to invoke some kind of act of baptism. What this does is to tie the name to a particular individual as that individual and not to a particular description, even in cases where the dubbing proceeds via description rather than ostension. This then allows for the possibility of changes, even fundamental changes, in the initial description. 29 On the other hand, it is through description that we learn to use these names effectively, in particular, when we are not acquainted with the bearer of the name. Thus, a description may play an epistemic role in distinguishing an individual from others, in terms of some set of properties, and in helping to fix reference through baptism, but it is not tied to the name by necessity. 30 As Gracia notes, this latter point can be taken as the essential insight of Kripke's discussion: we may use a description to get a fix on the bearer of 'Einstein', but once that fix is established we can change the components of the description through the construction of different possible worlds.
end p.208
Consider, yet again, the example of identity statements: on Gracia's account, the function of the names 'Morning Star' and 'Evening Star' is solely referential; in particular, there is no descriptive aspect to this function. 31 How, then, do such statements convey information? We learn to use the names 'Morning Star' and 'Evening Star' via descriptions involving both luminosity and spatio-temporal coordinates (for example, 'that bright object that appears at these positions in the sky on certain mornings/evenings of the year'). The information that we gain is that despite their differences, these are, in fact, descriptions of the same object. 32 That we are able to gain such information has nothing to do with the referential function of the names, but solely and appropriately with the distinguishing function of the descriptions. It is because we learn to use the names by means of this distinguishing function that we fall under the false impression that this function confers meaning upon the names; "[b]ut once it is understood that this is only a result of how we learned to use the names, then the difficulty vanishes". 33
What about existential statements? Here again we must carefully separate the referential and distinguishing functions. Take the statement 'Max Planck exists'. The function of 'Max Planck' is to refer and we intend to do so when we invoke it. However, the reference may fail, of course. Consider the lessons of history: we learn to use the name via a description-such as 'the first person to quantize energy'-and, following the causal chain, we assume there was some baptismal event at the beginning of it. But it may turn out that there was no such event, that the name is an invention, say.
What fictional names refer to, Gracia suggests, are possible individuals. 34 Thus 'Victor Jakob' refers to such a possible individual and statements such as 'Victor Jakob does not exist' and 'Victor Jakob could not accept the new quantum physics' are not absurd or meaningless, even though the function of 'Victor Jakob' is to refer and not to take the place of a description. 35 Of course, what is then required is further elaboration of this notion of possible individuals but we shall not pursue this here. Again, whether such statements mean anything or not does not presuppose that the entities actually exist; the meaning is cashed out descriptively but the name is not taken to be reducible to this. Gracia's claim is that with this separation between referential and descriptive functions, 'Max Planck was the person who first regarded energy as quantized' is no longer tautologous and we can accommodate the possibility of different people associating different descriptions with the same name:
. if . descriptions are only contingently related to individuality and serve to identify the bearer of the name only under very specific circumstances, then no problem arises by saying that proper names have no meaning and likewise for accounting for disparities of description. 36
What this sort of account suggests is that the assertion that what names do is enable us to isolate and maintain contact with individuals 37 is too crude. The latter concerns distinguishability, rather than individuality itself, and here, as Gracia points out, a description may be invoked but is not necessarily tied to the individual.
With this discussion behind us, we can now address three broad issues concerning the role of names in physics. First, it has been claimed that naming practices in physics support some form of Descriptivist view rather than the Causal or 'hybrid' theories. Secondly, it has been argued that Kripke's theory of rigidly designating names and the construction of possible worlds is tied to classical statistics and finds itself in difficulties in the quantum domain. Thirdly, however, the non-classical indistinguishability of quantum particles presents problems for both the Descriptivist and Causal approaches.
This discussion will then set the scene for an outline of Dalla Chiara's and Toraldo di Francia's quaset theory which we shall present as a theory of quantum individuals for which distinguishing descriptions are unavailable.
|
