Predicates and Properties

We are warming up to an examination of deflationary theories of truth according to which truth is either not a property or not a metaphysically substantive property.  (I oppose deflationary theories of truth just as I oppose deflationary theories of existence.) But first some clarification of 'predicate' and 'property.'

1. I begin by resisting the traditional conflation of predicates and properties, a conflation in evidence when we hear a philosopher claim that "existence is not a predicate."  That claim makes no sense unless a predicate is a property.  After all, 'existence,' as an abstract substantive, is not grammattically tuited to occupy predicate position.  If, however, a predicate is a bit of language used to express a property, then the claim should be that " '. . . exists' is not a predicate."  That's in order, as is "Existence is not a property."  As expressing properties, predicates are distinct from properties.  Predicates are linguistic while properties are extralinguistic.

To be a bit more precise, predicates (whether types or tokens) are tied to particular languages whereas the properties they express are not so tied.  Thus schwarz is tied to German in the way black is tied to English, but the property of being black is tied to neither.  Equally, the property of being disyllabic is tied to no one language even though it is a property that only linguistic items can have.  Thus 'Boston' but not Boston is disyllabic.

2. Some of you will question whether there are properties distinct from predicates.  Question away.  But just realize that in order to raise this very question you must first have distinguished predicates and properties.  You must already have made the distinction 'at the level of intension' if not 'at the level of extension.'  For you cannot maintain that there are no properties distinct from predicates unless you understand the term 'property' just as you cannot maintain that there are no unicorns distinct from horses unless you understand the term 'unicorn.'

3. By my lights, you are a very foolish philosopher if you deny properties, but not if you deny universals.  If you deny universals you are merely mistaken.  So let's be clear that 'property' and 'universal' are not to be used interchangeably.  It is a substantive question whether properties are universals or particulars (as trope theorists maintain).  Universals I define as repeatable entities, particulars as unrepeatable entities.

4. The predicate/property distinction under our belts, we need to note three views on their relation.

5. One view is that  no predicate expresses a property.  I rejected this view in #3.  To put it bluntly, there is a real world out there, and the things in it have properties whether or not there are any languages and language-users. Some of our predicates succeed more or less in expressing some of these properties.

6. A second view is that every predicate expresses or denotes a property.  The idea is that for every predicate 'P' there is a property P corresponding to 'P.'  But then, given that 'exists' and 'true' are predicates, it would follow straightaway that existence and truth are properties.  And that seems too easy.  Deflationists, after all, deny for reasons that cannot simply be dismissed that truth is a property.  They cannot be refuted by pointing out that 'true' is a predicate of English.  The following equivalence is undeniable but also not formulable unless 'true'  is a predicate:

'Grass is green' is true iff grass is green.

The deflationist will take an equivalence like this to show that 'true' is a dispensable predicate and therefore one that does not pick out a property.  (On Quine's disquotationalism, for example, 'is true' is a device of disquotation: it merely undoes the semantic ascent displayed on the LHS of the biconditional.)  We should therefore be uneasy about the view that every predicate expresses or denotes a property.  The existence of a predicate does not show the existence of a corresponding property.  A predicate need not predicate a property.  It should not be a matter of terminological fallout that wherever there is a predicate there is a property.

7.  Determined to maintain  that every predicate expresses or denotes a property, a deflationist  could of course hold that existence and truth are properties, but not metaphysically substantive properties.  A deflationist could argue like this:

Every predicate expresses a property

'True' is a predicate

Ergo: Truth is a property, but not a substantive one.

But he could also argue like this:

Every genuine predicate expresses a substantive property

Truth is not a substantive property

Ergo: 'True' is not a genuine predicate.

8.  A third view about the predicate-property relation has it that some predicates pick out properties and some don't.  I suggest this is how we should use 'predicate.'  It then becomes a matter of investigation, not of terminology, whether or not there is a property for a given predicate.

Comments

[Patrick]

This got me thinking about Max Stirner and his seminal work "The Unique One and His Own". An old friend of mine who was brilliant philosopher and philologist, and was fluent in German, claimed it should translate as "The Unique One and His Property". At the time of his early death we were exploring the idea of doing a new translation of this great work.

Have you studied Stirner? Any thoughts?

http://i-studies.com/library/the_individual_and_his_own/index.shtml

By the way, I like your blog. I have made it a regular stop in my day to day surfing.

Patrick

[William]

Just a pedantic nitpicky point here. I don't know where the maxim 'existence is a predicate', but it almost certainly originated in the period of traditional logic, according to which every proposition consists of just two terms, subject and predicate, joined by the 'copula'. The question was whether the verb 'is' simply unites subject and predicate, saying that Socrates = a white thing, for example. Or whether it genuinely is part of the predicate, or can be expressed as part of it.

In this traditional sense, read 'predicate' as 'property'. Although a further complication is that in Aristotelian logic, 'property' is one of the 5 predicables: genus, species, differentia, property, accident

In the modern sense, a predicate is simply a bit of a sentence with a gap in it. There can be more than one gap, e.g. '- loves -'. In the modern sense, obviously '- exists' has to be a predicate.

[Bill Vallicella]

Patrick,

Thanks for the comment and the kind words. I have the 1907 translation of Der Einzige und sein Eigenthum by Steven T. Byington as The Ego and his Own. My marginalia suggest that I read parts of it in 1973-74. But I didn't make a proper study of it, and haven't returned to Stirner since then. But of late I have been reading some anarchist stuff and this makes me want to re-read Stirner.

"The Unique One and his Property" would be a good translation of the title. Eigentum is what one has or possesses, what one owns, what is one's property. One would use Eigenschaft, however, if one were referring to an attribute or feature. Thus my car is part of my Eigentum whereas my weight is one of my Eigenschaften.

There are analogies between the two uses of 'property.' Thus we speak of possession in both cases. I possess my being male and I possess my car, but not in quite the same sense. I instantiate my properties in the Eigenschaft sense but not my properties in the Eigentum sense.

[Bill Vallicella]

W,
>>In this traditional sense, read 'predicate' as 'property'. << That was one of the points I made. But I was thinking of Kant: Offenbar, das Sein ist kein reales Praedikat. "Obviously, Being is not a real predicate." That means: Beng or existence is not a real property, where 'real' means: pertaining to quiddity or whatness.

>>In the modern sense, a predicate is simply a bit of a sentence with a gap in it.<< Right. Start with 'London exists,' take out 'London' and you get: ' ___ exists.' So it is obvious that 'exists' is a first-level grammatical predicate. But it doesn't follow that existence is a first-level property. In a logically sanitized language a la Frege, 'exists' is not an admissible first-level predicate.

[Patrick]

Steven T. Byington

Thanks for that. I thought Tucker was the first.
http://en.wikipedia.org/wiki/Benjamin_Tucker

[Komal]

Thanks for writing this, I learned a lot from it.

[Bill Vallicella]

You're welcome, Komal. Part of the purpose of this blog is to provide free philosophy lessons.