Sunday, December 18, 2011

Early Moderns and "Thinking Around"

I am taking a brief break from grading to make a few notes about something I've become increasingly interested in recently, which I've been labeling for myself as "thinking around" (to be contrasted with "thinking about").

I'm going to start with two examples, one from Hume and one from Berkeley.

On my reading of Hume, there is a sort of mental activity one can engage in towards that which is strictly and literally inconceivable.  This activity is supposition.  In one part of my dissertation, I attempt to show that Hume can embrace this form of mental engagement without abandoning his commitment to analyze all mental activity (of the understanding) in terms of conception (i.e. ideas).  At any rate, there are a few passages which are naturally read as Hume allowing that some things can be supposed which cannot be conceived.  This type of mental engagement, I argue, allows a response to a Reidian objection which charges Hume as unable to account for reductio ad absurdum reasoning.  So, while you cannot, on my reading of Hume, think of or about an even prime greater than 2, for example, you can think around such a prime, allowing you to reason your way to its non-existence.

Berkeley, like Hume, has a view of conception bound up with what ideas one possesses.  Consequently, Berkeley deploys arguments about the nature of ideas to show that certain things are inconceivable.  But, as is somewhat explicit in the third Dialogue between Hylas and Philonous, and fully explicit in Alciphron VII, Berkeley introduces a way to defend the meaningfulness of discourse in which meaningful terms to not signify ideas (rejecting a straightforward Lockeanism about language), with something I'll call "notions" (though I don't know if Berkeley consistently uses the "idea"/"notion" terminology to track this distinction).  Having a notion of something does not require having an idea of it.  Thus, even though I cannot have an idea of immaterial susbtance, I still have a way to engage with propositions about immaterial substances (whether we are speaking of me or god).  This too is a sort of thinking around, as I understand it.

Resources which allow a philosopher to permit our thinking around something which we cannot (on their view) properly think about or of are important elements of their views for two reasons.  First, they can give us important insights about other aspects of their views.  For instance, noting that Berkeley must appeal to some such resource in the third dialogue, to explain how we can believe in immaterial substance helps us exclude some (seemingly natural) interpretations of the first dialogue arguments against material substance.  While it might appear that Berkeley is offering a straightforward inconceivability argument against belief in material substance there, it is clear from his own later admission that we cannot strictly conceive of immaterial substance that the dialogue one argument must be more complicated than it at first seemed.

Second, however, they are important for allowing us to see how powerful objections to those philosophers wind up being.  Take Hume, who embraces the view that we cannot conceive of anything which is impossible.  Given that various philosophers have appeared to sincerely defend views which, for Hume, turn out to be impossible, there is the objection that Hume cannot be right, because we could not then make sense of such apparently sincere defenses.  A natural sort of reply is to invoke some sort of verbal confusion underlying the dispute.  But that line of reply is not always satisfying, and does not always do a good job of addressing the behavior of his opponents.  On the other hand, Hume's resource of supposition-without-conception permits him a more robust way to understand his opponents as engaging with these impossible views (apart from merely "mistakenly defending that the sentences which express those impossibilities actually express truths").

I'm sure that similar sorts of resources crop up in the views of other philosophers, but I don't want to just start casting around randomly. If anyone has suggestions of places to look (especially in terms of early modern figures other than the "canonical" British empiricists), please let me know.

Tuesday, December 6, 2011

Monday Mill Blogging (#006)

Monday Mill blogging makes its triumphant return to Tuesdays, apparently.

Today's post covers book 1, chapter 2, section 3.

§ 3.  General Names and Singular Names

Section §3 is largely focused on the distinction between general and singular names.  Mill's infamous doctrine about singular names won't be put forward until two sections later, but there are some interesting elements to this discussion nonetheless.  The section opens:
All names are names of something, real or imaginary; but all things have not names appropriated to them individually. For some individual objects we require, and consequently have, separate distinguishing names; there is a name for every person, and for every remarkable place. Other objects, of which we have not occasion to speak so frequently, we do not designate by a name of their own; but when the necessity arises for naming them, we do so by putting together several words, each of which, by itself, might be and is used for an indefinite number of other objects; as when I say, this stone: "this" and "stone" being, each of them, names that may be used of many other objects besides the particular one meant, though the only object of which they can both be used at the given moment, consistently with their signification, may be the one of which I wish to speak. (p. 27)

First off, I am not sure, but it sounds like Mill might be going in for a quasi-Meinongian view, given that opening claim.  There might be a way to cash it out that doesn't involve quantifying over non-real entities, but it seems like Mill is suggesting that, while not every object has a proper name, every name has a proper object.  This sort of commitment is relevant to evaluating, for instance, what we can take Mill as having to say, if anything, about Frege's puzzle and/or other puzzles for the Millian view of proper names.  For instance, if Mill is positively committed to imaginary objects as being denoted by meaningful terms, this provides him with something of a response to at least one problem arising from "empty" names.  Actually, it might be apt to call it a "presponse", since it looks like Mill recognized that not all meaningful names correspond to real objects, and offered a view about those cases antecedent to a specific challenge being issued.

So, Mill thinks that putting together general names into complexes can supply us (perhaps only incidentally and temporarily) with names for objects that don't have their own proper name.  This is a pretty plausible thesis, I'd say.  However, this is not the only purpose for general names.  Mill suggests that if all general names were for is to allow ad hoc construction of names for objects that don't have their own proper names, they "could only be ranked among contrivances for economizing the use of language".  There is a parallel here between Mill and Locke on the role of general terms/names.  Both recognize that a language where every objects possesses only a proper name would be unmanageable, and accept some sort of argument from the practical necessities of language in favor of general terms.

As noted, Mill does not think this is the only role of general terms:
But it is evident that this is not their sole function.  It is by their means that we are enabled to assert general propositions; to affirm or deny any predicate of an indefinite number of things at once. The distinction therefore, between general names, and individual or singular names is fundamental; and may be considered the first grand division of names. (p. 27)

Mill then defines a general name as a name that is capable of being truly affirmed, in the same sense, of each of an indefinite number of things.  Individual or Singular names, then, are capable of being affirmed, in the same sense, of one thing only.  Mill's example of a general name is "Man" and his example of a singular name is "John".  Mill is not concerned about the fact that lots of people are named "John", because he thinks that "John" in "John Lennon" has a different sense than "John" in "John Fitzgerald Kennedy".  Here is the interesting part.  In defending this claim, we get a preview of Mill's famous claim about proper names:
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. (p. 28)

I am not sure how to make sense of this claim made here, that "John" cannot be affirmed of people named "John" at all, and the earlier claim that "a singular name is a name which is only capable of being affirmed, in the same sense, of one thing."  The definition seems to suggest that singular names are affirmed of individuals, while the latter remark seems to suggest that (some) singular names are not affirmable at all.  Now, Mill goes on to say that "The king who succeeded William the Conquerer" is a singular name, and presumably that can be affirmed, so some singular names would still be affirmable.  However, Mill's point that the propriety of calling someone "John" does not depend on their antecedently possessing some feature that is designated by "John" seems right.  Intuitively, the dependence goes in the other direction, the quality of "going by the name 'John'" is had in virtue of being called "John".  Mill's reason for taking "The king who succeeded William the Conquerer" to be singular is this: "that there cannot be more than one person of whom it can be truly affirmed, is implied in the meaning of the words."  In other words, there is a sort of semantic guarantee of the term applying to at most one object, and this suffices for it to be singular.

I have not mentioned revisions or amendments to the text, though the Liberty Fund edition of the text I am using has ample detail about changes to the text between the manuscript and different editions.  I do want to make mention of an interesting revision to the passage I've just been discussing.  The earlier text had, as the example definite description, "The present King of England" and in the explanation of it qualifying as singular, he said, "never can be more than one person at a time of whom it can be truly affirmed".  This revision is interesting to me because I think Mill would still want to count "The present king of England" as a singular name, but it seems that it is a messy example to use, since it can, at different times, be affirmed truly of different people.  I suppose one could say that it is being used in different senses at different times, but then to explicate this, one would have to suggest that the word "present" undergoes a continual change of sense.  While this might, ultimately, be the best thing to say about it on Mill's view, it would make things much messier to lay all this out when trying to explain the division than to treat of the quirkiness of the example later, when more of the machinery is in place.

Back to the main text: Mill observes that even an incomplete definite description, such as "the king", can, in the right context, count as an individual name.  Like the point about "the present king of England", this looks to open the door to all sorts of complications, at least if one tries to reconcile the official definitions of singular and general names with a willingness to allow context to dictate the singular/general nature of a name.

The last two things Mill does in this section are: a) complain about use of the word "class" to define "general name" and b) distinguish between collective singular names and general names.

On (a):  Mill says that it is common for people to define general names by saying general names are names of classes.  "But this, though a convenient mode of expression for some purposes, is objectionable as a definition, since it explains the clearer of the two things by the more obscure."  Mill goes on to propose that the definition be reversed, seemingly insensitive to the fact that this would rule out unnamed classes.  I don't know if that is a major issue, but it seemed worth observing.

On (b): Here Mill is essentially telling us that general names are predicated distributively, collective names predicated jointly.  "The 76th regiment of foot in the British army" is a collective singular name for a group of soldiers.  There is only one group (at a given time) of whom you can properly affirm that name, and you can't affirm it of each of the individual members.  "Regiment" is Mill's example of a collective general name, since it can be affirmed of a lot of different groups in the same sense.  Mill suggests that it is "general with respect to all individual regiments, of each of which separately it can be affirmed: collective with respect to the individual soldiers of whom any regiment is composed."  This last line suggests that collectivity is type-relative.  This is good, because it means we don't have to decide all questions of collectivity in our basic semantics.  "Mt. Everest" can be non-collective with respect to the category mountain, but still turn out to be collective with respect to the category particles of matter.

Next time on Monday Mill Blogging: §4, "Concrete and Abstract Names"

Thursday, December 1, 2011

Monday Mill Blogging (#005)

Monday Mill blogging on a Thursday? Why not.

Today's post covers book 1, chapter 2, section 2.  All this focus on naming is making me want to take some time to re-read part III of Carnap's "Meaning and Necessity".  But for now I am sticking with the Mill.

§ 2.  Words Which are Not Names, but Parts of Names

Mill ended §1 by indicating the need to outline a taxonomy of names.  But before he will give us his taxonomy of names, he feels it is necessary to discuss words that are not properly considered names, but which are parts of names.  Mill shares the conventional wisdom of which words those are:
Among such are reckoned particles, as of, to, truly, often; the inflected cases of nouns substantive, as me, him, John's; and even adjectives, as large, heavy.  These words do not express things of which anything can be affirmed or denied.  We cannot say, Heavy fell, or A heavy fell; Truly, or A truly, was asserted; Of, or An of, was in the room. Unless, indeed, we are speaking we are speaking of the mere words themselves, as when we say, Truly is an English word, or, Heavy is an adjective. (p. 25)

Mill's view here seems to be that words, in addition to their customary uses, can be used to denote "the mere letters and syllables of which [they are] composed", and in that usage, words like "of" and "heavy" are names.

Ultimately, Mill is going to remove adjectives from this list, and treat them as names.  He explains his reasoning as related to the fact that it is a mere grammatical accident of English that we cannot say "A heavy fell".  Mill marshalls some cross-linguistic evidence from Greek and Latin in support of this point, and then reaffirms that adverbs and particles can't ever denote terms in a proposition (except when being used as names for the words themselves).

Mill then puts the views he has just been outlining in scholastic terms.  What he is calling names are what the scholastics called Categoremic terms, the words that are not names, but only parts of names, are the scholastics' Syncategoremic terms.  Rather than have a third class for compound terms ("A court of justice"), Mill treats these as many-word names, and classes them as Categoremic.

In treating of these many worded names, Mill also presents a view on non-restrictive relative clauses (though he doesn't call them that):
Thus, when we say, John Nokes, who was the mayor of the town, died yesterday—by this predication we make but one assertion; whence it appears 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 assertion, 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." (p. 27)

I say this is a view, even though it seems a bit cursory in terms of detail, because it, in a sense, helps us figure out what Mill would want to say about the truth or falsity of a sentence of the form, "n, who was G, is H", when the referent of 'n' has the property designated by 'H', but not the property designated by 'G'.  The use of that sentence, it seems, makes the proposition, of the referent of 'n', that they have the property designated by 'H', so the primary assertion made in uttering the sentence is true.  However, in the subject term of the sentence "there is included another assertion", the assertion, about the referent of 'n', that they possess the property designated by 'G', which is false.  It isn't clear whether this gives us a satisfactory answer about how to classify the sentence "n, who was G, is H", relative to a circumstance of evaluation, but it does shed some light on how Mill thinks about the relationship between sentences and assertions.

Next time on Monday Mill Blogging: §3, "General and Singular Names"