Mill: names meaning things

1. Taking things in some sort of historical order, we shall now look at John Stuart Mill's account of meaning, as given in Book I, 'Of Names and Propositions', of his System of Logic, 1843 Mill exemplifies the "psychologism" that so enflamed Frege. In the Introduction he defines Logic as "the science of the operations of the understanding which are subservient to the estimation of evidence: both the process itself of advancing from known truths to unknown, and all other intellectual operations in so far as auxiliary to this". His claims following this about how far he will take the analysis show that he was not really interested in mental activities as such, rather than the objective contents of such activities and their logical relations. Book I begins with some reasons for the prior study of language: most thought is in words; you cannot investigate the "import of Propositions" without attending to words, and propositions are the objects of all belief and knowledge. Mill's first claim about propositions/objects of belief is that they are formed "by putting together two names". A proposition is "discourse, in which something is affirmed or denied of something", so Mill goes on immediately to say that "every proposition consists of three parts: the Subject, the Predicate, and the Copula". He argues that nothing can be believed or can be true or false unless it includes at least this much structure. You can think of or conceive something when you hear the words 'the sun' but you cannot yet believe anything. There is then a big difference between 'the sun' and 'the sun exists', and in particular the former does not mean what the latter says. He here uses the example of 'a round square' to show that this phrase does not, and cannot mean 'a round square exists' since it is impossible that such a thing should exist.
2. To get beyond this simple starting point Mill thinks we need to look at what Names do ( a capital letter when using the word in Mill's extended sense outside quotations from him: 'yellow' would not normally be called a name in ordinary English, but it is a Name for Mill in 'gold is yellow').
3. Chapter II begins by quoting Hobbes saying very much the same as Locke about words as standing for thoughts and permitting us to communicate our thoughts to others. Mill comments that this claim "appears unexceptionable. Names, indeed, do much more than this; but whatever else they do, grows out of, and is the result of this". But he then goes on to ask whether Names are better seen as standing for things or our ideas of things. Taking Hobbes' claim that the word stone is a sign of a stone only in the sense that it tells us that its user is thinking of a stone, Mill comments that

If it be merely meant that the conception alone, and not the thing itself, is recalled by the name, or imparted to the hearer, this of course cannot be denied. Nevertheless, there seems good reason for adhering to the common usage, and calling (as indeed Hobbes himself does in other places) the word sun the name of the sun, and not the name of our idea of the sun. For names are not intended only to make the hearer conceive what we conceive, but also to inform him what we believe. Now, when I use a name for the purpose of expressing a belief, it is a belief concerning the thing itself, not concerning my idea of it.

'The sun is the cause of day' means something about the physical consequences of the sun's presence, not that my idea of the sun makes me think of day. Mill says that he will therefore take Names to stand for things themselves, not merely for our ideas of them. Mill starts out with a very rich conceptual structure: there are names and they “get at” things and we can “get at” what they get at without the name to verify the connection.

4. Before considering the varieties of Names, Mill recognizes that not all words are Names, some are "only parts of names". Among such are thought to be particles (e.g. 'of', 'often'); inflections (the apostrophe “s” of 'John's'); and even adjectives (e.g. 'large'). "These words do not express things of which anything can be affirmed or denied. We cannot say, Heavy fell". Mill notes what mediaeval logicians called suppositio materialis (mention) when we use 'of' to mean the word 'of', as in 'Of is an English word'. But this is clearly a special case. But Mill then rejects the doctrine as applied to adjectives since he recognizes that adjectives can stand by themselves as the predicate of propositions ('snow is white') and can even serve as subjects ('white is an agreeable color'). So adjectives will be regarded as Names. In terms of traditional categories Mill takes Names to cover nouns, adjectives and most verbs.
5. Mill notes another bit of mediaeval terminology, for the words that can only operate as parts of a Name: syncategorematic since they must work with [syn-] a word from one of the basic categories. Mill then goes on to offer a test for deciding how many Names we have when we use a string of words. Take such a string, predicate something of it, and observe whether that makes one assertion or several. His example is worth considering for comparison with later writers. In the sentence

John Nokes, who was the mayor of the town, died yesterday.

Mill tells us that his test shows that 'John Nokes, who was the mayor of the town' is no more than one Name.

It is true that in this proposition, besides the assertion that John Nokes died yesterday, there is included another insertion, namely, that John Nokes was mayor of the town. But this last assertion was already made: we did not make it by adding the predicate, "died yesterday." Suppose, however, that the words had been, John Nokes and the mayor of the town, they would have formed two names instead of one.

6. "All names are names of something, real or imaginary". We don't have Names for every object, only for particulars of some standing significance for us. When we don't have a proper name, we designate a particular object by combining words, as with 'this stone'. Each word could be used to stand for any number of things; put together on a particular occasion the phrase designates one stone. But designating particulars is not our only desire; we want to assert general propositions. So the first divison of Names is into general and individual or singular. A general name can be affirmed of each of a number of things; an individual name can only be "truly affirmed, in the same sense, of one thing". 'Man' is general since it can be affirmed of John, Mary, etc. in the same sense,

for the word man expresses certain qualities, and when we predicate it of those persons, we assert that they all possess those qualities. But John is only capable of being truly affirmed of one single person, at least in the same sense. For, though there are many persons who bear that name, it is not conferred upon them to indicate any qualities, or anything which belongs to them in common; and cannot be said to be affirmed of them in any sense at all, consequently not in the same sense.

7. Mill next distinguishes general from collective Names. A general name can be predicated of each individual of a multitude; a collective name cannot be predicated of each separately. Compare 'soldier' with 'the 76th regiment of foot in the British army'. The latter is a collective individual Name.
8. The next main division of Names is between concrete and abstract. Concrete terms stand for things; abstract ones for attributes of a thing. 'Old' is concrete; 'old age' is abstract. Mill acknowledges the scholastic origin of his distinction and notes that Locke has muddied the waters by talking of all general terms as abstract. He raises but wishes to avoid a definitive answer to the question whether abstract Name are general or singular. He also considers an objection that words like 'white' stand for colors, that is attributes. His answer is that when we say 'snow is white' we intend to say that certain things have a color, not that they are a color; so 'white' is a Name of all things that are white; the attribute 'whiteness' is part of the word's signification since the word implies the attribute, but it is not a name of the attribute.
9. These ideas lead to the next major distinction: connotative and non-connotative Names.

A non-connotative term is one which signifies a subject only, or an attribute only. A connotative term is one which denotes a subject, and implies an attribute.

Examples of Millian non-connotative Names are: London, John, whiteness, virtue. Connotative Names include 'white' and 'virtuous'.

The word white, denotes all white things, as snow, paper, and the foam of the sea, &c; but when we predicate it of them, we convey the meaning that the attribute whiteness belongs to them.

Here we see Mill's contrast between what a Name denotes and what it connotes; the whole range of entities of which it is correctly predicated versus the attribute in virtue of which those predications are true.

10. Mill observes that all concrete general Names are connotative. Given what 'man' connotes, a race of intelligent language-using elephants would not be called 'men', nor probably would animals with the form of men but with no vestige of reason. Some abstract general Names are also connotative; Mill offers 'fault' meaning a bad or hurtful quality.
11. Proper names are not connotative. There may be a reason why we give someone a particular proper name, but Mill's point is that that reason is no part of what the name then means. John may have been named for his father; Dartmouth may be a town at the mouth of the river Dart. But in neither case is this part of the signification of the word. "Proper names are attached to the objects themselves, and are not dependent on the continuance of any attribute of the object."
12. Mill then notes a kind of individual name that is connotative, where the words may "be significant of some attribute, or some union of attributes, which, being possessed by no object but one, determines the name exclusively to that individual". His main examples here are of what have become known as definite descriptions: the father of Socrates, the first emperor of Rome, the present Prime Minister of England.
13. When words connote, their meaning is what they connote, not what they denote. "The only names of objects which connote nothing are proper names; and these have, strictly speaking, no signification." Mill likens the use of proper names to making a chalk mark on a house to allow you to pick it out again from its indistinguishable neighbors. We make a mark, not upon the object, but the idea of the object. This allows us to attach predications to the particular "thing with which we were previously acquainted".
14. We learn a proper name by learning to whom it is attached, its denotation. But we cannot thus learn connotative Names. A large number of different connotative Names will all denote the same individual. We often learn the denotation of a phrase before we grasp its connotation: Mill's example is that a child will know its brothers and sisters before knowing "the facts involved in the signification of those words".
15. Sometime connotation is imprecise. We may not have had to make a decision. This explains his uncertainty about non-rational man-like creatures. We haven't had to decide how similar to us a new kind of animal must be to be called a man. Such imprecision can be useful, but we need to be vigilant. Too often we use words with little idea of their connotation, having merely made a rough guess from the examples of their denotation we have come across.

In this manner, names creep on from subject to subject, until all traces of a common meaning sometimes disappear, and the word comes to denote a number of things not only independently of any common attribute, but which have actually no attribute in common; or none but what is shared by other things to which the name is capriciously refused.

Mill gives an example from Bain's Logic in a footnote:

Take the familiar term Stone. It is applied to mineral and rocky materials, to the kernels of fruit, to the accumulations in the gall-bladder and in the kidney; while it is refused to polished minerals (called gems), to rocks that have the cleavage suited to roofing (slates), and to baked clay (bricks). It occurs in the designation of the magnetic oxide of iron (loadstone), and not in speaking of other metallic ores. Such a term is wholly unfit for accurate reasoning, unless hedged round on every occasion by other phrases; as building stone, precious stone, gall stone, &c. Moreover, the methods of definition are baffled for want of sufficient community to ground upon.

Mill goes on to mention the difficulties this slackness creates for careful philosophical discussion, noting the strident controversies caused by attempts to define terms that have been subject to this obfuscation.

16. Indeterminacy of sense is different from ambiguity: two or more Names that share the same word.
17. Mill ends this section with a contrast between his usage and that of his father. That is no longer a live issue for us, but it is worth noting that the Millian or philosophical usage of 'connote' is different from that of literary scholars and the general public. For them a word's connotation is something like the aura or atmosphere its use suggests, something, as my dictionary says, "in addition to the literal meaning". For Mill, connotation is the literal meaning.
18. Chaper II continues with some distinctions we do not need to examine. Mill's importance for us at the moment derives from his doctrine of connotation and denotation and the particular claim he made for the existence of a class of non-connotative names that serve simply to introduce objects into our thought.

Lycan's discussion

19. Lycan uses the phrase 'Millian name' for the view that there are words that are "merely labels for individual persons or objects and contribute no more than those individuals themselves to the meanings of sentences in which they occur" (p. 37).
20. Kripke is the source for the use of the term 'rigid' to stand for a characteristic feature of most ordinary proper names, and one that would be shared with Millian names if there are any: a rigid designator picks out the very same object in every possible world in which that object exists. Suppose George Bush Junior is President of the USA. As it happens, 'the President who finally ousted Saddam Hussain' refers in this world to George Bush Junior. In another possible world, it may have referred to his father, and in yet others to Bill Clinton. The name, George Bush Junior, however, refers to the same man in all such fantasies. It is rigid.
21. Lycan observes (pp. 55-6) that not all rigid designators are Millian names. 'The positive square root of nine' may well be rigid, but it is most certainly not a Millian name. Mill would have said it was a connotative individual name, as we have seen.
22. Whatever Kripke himself believes, many have recently defended theories of ordinary proper names that treat them as Millian. As Lycan notes, this view is often called the "Direct Reference" theory of names.
23. Lycan confronts all theorists of proper names with four puzzles.
1. Apparent Reference to nonexistents
   We saw Mill noticing that some Names are names of something imaginary. But what precisely is something imaginary?
2. Negative existentials
   A special case of the preceding: 'Vulcan does not exist' is true and seems to be about Vulcan; but if it is true there is nothing for it to be about. If there is something for it to be about, then it is false!
3. Frege's puzzle about identity
   'Clark Kent is Superman' seems more informative than 'Clark Kent is identical with Superman'. But if proper names are Millian then these two sentences have the same meaning.
4. Substitutivity
   There seem to be many sentences in which you cannot substitute co-referring names and preserve truth-value. Quine called such contexts 'referentially opaque'. Standard examples arise when talking about beliefs:

John believes Cicero denounced Catiline.

may be true, while

John believes Tully denounced Catiline.

may be false, despite the fact that the two names 'Cicero' and 'Tully' are names of the same man. If these names are Millian, this couldn’t happen?

24. Lycan notes that Direct Reference theorists have challenged the argument from opaque contexts by trying to show that there are non-opaque readings of such sentences. While this may be so, he makes the point that the Direct Reference people need more than that; the Millian has to say that there aren't opaque contexts and this seems too hard to swallow. He leaves us at this point with a trilemma: the Millian can't account for the puzzles; nor can the views that derive from Frege; nor can those that derive from Russell.

Searle’s Objection

25. Searle argues that it doesn’t make sense to treat names as “short for” truncated descriptions and not Millian names.
26. Suppose we say,
1. Wilfrid Sellars was an honest man
27. Russells analysis commits him to the view that whatever the uniqueness condition logically entails, the speaker must mean which seems bizarre in the extreme. Rarely do speakers “Have in mind” a particular description and, often enough, there isn’t any. If Wilfrid Sellars is equivalent to “The one and only thing x such that x is F ans x is G and ….,” for all predicates, then I would have in mind whatever satisfied those predicates. Lycan gives the example of an argument with Sellars (that did happen by the way).
28. Objection 2
1. The Name Claim devolves names into multiply ambiguous terms which does not seem right.
2. As a result, we may not be able to disagree about a person. Lycan gives two examples (42) to illustrate the point. We can argue past each other but it is unlikely that we will actually pick out the same person.
3. Recall, the analysis is
1. (3x)(Qx & ((y)( Qy à y=x) & x = ws)) where each of us might have a different property for Q.
29. Searles’ Cluster theory
1. Can Russell’s analysis be saved by introducing a more open-ended referent for the name? He argues for a cluster theory that postulates a sufficient but vague unspecified number of identifying statements (SBVAUN).
2. As Searle’s response careens off into unintelligibility, Kripke broadsides him with an argument against any account that relies on descriptions.

Kripke’s Critique

30. Modal contexts are common so it seems right that a theory should handle them.
1. Nixon won the 1968 US Presidential election. Could he have lost? The answer seems obviously “Yes” but how does Russell’s theory handle it?
1. On the description theory this is equivalent to
2. (11) It is possible for RN to have lost the 1968 election

Which is scope ambiguous and can be either the scope

<> ((3x)(Wx & (y)(Wy à y=x) & (z) (Wz à ~Wz)))

Which is false and the scope reading

((3x)(Wx & (y)(Wy à y=x) & (z) (Wz à <> ~Wz)))

Which makes is true.

Objection 6

31. According to the Name Claim, every name can be cashed in for a unique description. But most people associate “Cicero” say, with “a famous Roman orator.” It seems like pretty thin cash but, as a practical matter, people succeed in referring anyway. This shouldn’t be possible according to the Name Claim. It seems outlandish to say that in such cases people can refer successfully.

Objection 7

32. Russell wanted his theory to apply to fictional names. According to the Name Claim, any sentence containing a fictional name will come out false. But some fictional sentences, our intuitions tell us, are true. The Tooth Fairy, for example, isn’t fat, doesn’t have a beard or work at the North Pole.