Nominalism and Being

Today I preach on an old text of long-time commenter and sparring partner, London Ed:

Nominalism is the doctrine that we should not multiply entities  according to the multiplicity of terms. I.e., we shouldn't  automatically assume that there is a thing corresponding to every  term. Das Seiende is a term, so we shouldnât automatically assume there is a thing corresponding to it. Further arguments are needed to show that there is or there isnât. A classic nominalist strategy is to rewrite the sentence in such a way that the term disappears.

 My first concern is whether this definition of 'nominalism' is perhaps too broad, so broad that it pulls in almost all of us. Does anyone think that every term has a referent? Don't we all hold that there can be no automatic assumption that every occurrence of a term in a stretch of discourse picks out an entity? For example, one would be hard pressed to find a philosopher who holds that 'nothing' in

   1. Nothing is in the drawer

refers to something. (Carnapian slanders aside, Heidegger does not maintain this, but this is a separate topic about which I have written a long unpublished paper.) Following Ed's excellent advice, the
apparently referential 'nothing' can be paraphrased away:

   2. It is not the case that there is something in the drawer.

This then goes into quasi-canonical notation as

   2*. ~(Ex)(x is in the drawer).

In (2*) the tilde and the particular quantifier are syncategorematic elements. On the face of it, then, there is no call to be anything other than a nominalist about 'nothing,' using 'nominalism' as per the
suggestion above.

Whether there is call to be a nominalist about 'being' is another matter. Before proceeding to it, consider the following example:

   3. Peter and Paul are blond

which could be parsed as

   3*. Peter is blond and Paul is blond.

Now I rather doubt that anyone maintains that every word in (3*) -- or rather every word in a tokening of this sentence-type whether via utterance or inscription or some other mode of encoding -- has an entity corresponding to it. This suggests a taxonomy of nominalisms:

Mad-Dog Nominalism: No word has an existing referent, not even 'Peter' and 'Paul.' (I write 'existing referent' to disallow Meinongian objects as referents. The waters are muddy enough without bringing Meinong into the picture -- please pardon the mixed metaphor.)

Extreme Nominalism: The only words that have existing referents are names like 'Peter' and Paul'; nothing in reality corresponds to such predicates as 'blond.' And a fortiori nothing corresponds to copulae and logically connective words like 'and' and 'or.'

Nominalism Proper: Particulars (unrepeatables) alone exist: there are no universals (repeatables). This view allows that something in reality corresponds to predicates such as 'blond.' It is just that what this predicate denotes is not a universal but a particular, a trope say, or an Aristotelian accident.

Methodological Nominalism: This is just Ed's suggestion that we not assume that for each word there is a corresponding entity.

I hope no one is crazy enough to be a mad-dog nominalist, and that everyone is sane enough to be a methodological nominalist. The two middle positions, however, are subject to reasonable controversy. What I am calling Extreme Nominalism has little to recommend it, but I think Nominalism Proper is quite a reasonable position.  There has to be something extralinguistic (and extramenal) corresponding to the predicate in 'Peter is blond,' but it is not obvious that it must be a universal.  

Now let's think about whether we should be nominalists with respect to words like das Seiende, that-which-is, the existent, beings, and the like. Heidegger has been known to say such things as Das Seiende ist,  or

   4. That-which-is is. (Beings are.)

Now is there anything in reality corrresponding to 'that-which-is' and 'beings'? Well of course: absolutely everything comes under 'that-which-is.' There is nothing that is named by 'Nothing.' And if I met nobody on the trail, that is not to say that I met someone named 'Nobody.' But absolutely everything falls under 'a being,' 'an existent,' ein Seiendes, das Seiende.

So I see no reason to have any nominalist scruples about the latter expressions. I don't see any problem with forming the substantive das Seiende from the present participle seiend.  But you will be forgiven if you balk at the transformation of the infinitive sein into the the substantive das Sein and take the latter to refer to Majuscule Being.

Comments

[Daniel Aledo]

Good day Mr. Valicella. I tried to comment on the section "God, Possibility, and Evidential Support for Non-Contingent Propositions" but I am not allowed to post comments there and I cannot find your email address.

I apologize for posting here. I am having a discussion on evidential arguments from evil. I am using your insight that if God exists, then he necessarily exists and if God does not exist, he is impossible and because of this evidential arguments are irrelevant to the existence of god.

Now I am being attacked by several people saying that you can have evidence against an impossibility. one person said, for example,that if “the prior probability of a hypothesis is zero that does not mean that there can never be evidence against it.”

I am not sure how could I respond to this, are they making a fallacious argument? do they have a point? can you please lend me a hand?

[Bill Vallicella]

I wish I could help you, Mr Aledo. Unfortunately, this topic is 'above my pay grade.' I have spent some hours puzzling over how the alethic modalities interact with probability theory, but I am still in a fog about it. I think I have an unfinished post on this which I will take another look it and see if it is worth posting.

[Edward Ockham]

I like the classifications. Some comments.

1. On your definition of 'nominalism proper'. "Particulars (unrepeatables) alone exist: there are no universals (repeatables)." How do we define the notion of a 'repeatable'? If you mean, the same thing being repeated, that is contradictio in adiecto, for what is one thing cannot at the same time be many things. Or do you mean that there is some form of unity that is less than numerical unity? This would be Scotus' position, which Ockham repeatedly (!) and convincingly argues against. Ockham says that 'species' is not the name of any common nature existing in reality outside the mind, but is an 'intention in the soul'. Thus 'man is one species' does not predicate one universal of another. Rather, it predicates a second intention ('species'), which is is singular but exists in the mind only, of a first intention ('man') which also exists only in the mind.

2. On 'Methodological nominalism', you perhaps need to add to this the thesis that by some trick of language (as Wittgenstein would say), or by ignorance of logic (as Ockham would say) we are led astray into the false and fantastical belief that certain terms have a referent, when they do not, or when they have a quite different kind of referent. As an example (Summa I, c. 65), Ockham gives the proposition 'man is a species'. This has two senses, according as we take 'man' to stand for a mental concept, predicable of many, and according as it stands for all men. In the first sense it is true, for it says that one mental concept 'man' falls under mental concept 'species'. However, we are beguiled by the superficial resemblance to propositions like 'a man is talking', and we falsely imagine that it is being asserted that some thing signified by ‘man’ is a species, "which is manifestly false".

3. On your 'Mad dog' nominalism, I agree that it is distinct from 'extreme nominalism' of the Ockham variety (Ockham agrees that there are things referred to by proper names, and also agrees that qualities also have a real existence. But there is a further variety ('Londonism') which, while it does not deny the existence of Peter and Paul, denies the existence of a 'reference relation'. That is my position. Ockham is a realist about the 'supposition relation' which exists between the name 'Peter' and the person Paul. I am not a realist, for I deny the existence of any such relation.

[Bill Vallicella]

The 'repeatable'-'unrepeatable' terminology I believe I got from David Armstrong. It is the best was to explain the difference between universals and particulars. If you have ten red disks of the same shade of red, and 'red' picks out a universal, then redness is repeated in each of them -- the same universal occurs in each. Particulars are not repeatable in this way.

The idea is not that one thing is many things, but that numerically one and the same thing is present in or exemplified in many things. The idea is perfectly coherent as far as I can see.

As for the 'reference relation.' If you deny that there are universals, and if relations are universals, then you must deny the reference relation.

[Edward Ockham]

>> redness is repeated in each of them -- the same universal occurs in each

[my emphasis] Numerically the same universal? How can the numerically same thing be both singular and several? If not numerically the same thing, is it that these numerically different universals share some common feature in virtue of which they are the 'same' universal? What is this common feature? Is it one or several? If one, is it repeatable (continue argument ad infinitum).

[Bill Vallicella]

Ed,

I really don't understand your problem. I don't claim that there are no problems associated with the positing of universals. But you seem to think that the VERY IDEA of a universal is incoherent. That I do not see.

One and the same universal U is present in a, b, c, . . . x. Why do you think U must be several? What do you mean by 'several'?

[Edward Ockham]

>> But you seem to think that the VERY IDEA of a universal is incoherent.

No, I was confused by your manner of expression. You said "redness is repeated in each of them -- the same universal occurs in each". The verb 'occurs' strongly suggests that what occurs in one instance is numerically different from what occurs in another. 'Repeated' and 'is present in' have similar connotations.

If you mean that there is some object U and some relation R such that R(Socrates, U) and R(Plato, U) and so on, and that this relation holds in virtue of 'Socrates is a man' being true, and 'Plato is a man' etc, then that makes more sense. Perhaps. But even then we have to explain in virtue of what the predicate 'R(---, U)' applies to Socrates and to Plato, and I can't see a way out of the infinite regress.

[Bill Vallicella]

Ed,

There are more versions of realism than are dreamt of in your philosophy . . . .

Daniel,

Thank you for the kind words. I am glad my writings have been of use to you.

Here are some questions to think about. What is evidence? Is all evidence empirical evidence? What does 'empirical' mean? Is the question whether there are married bachelors an empirical question? Is empirical evidence relevant to answering it? Are arithmetical truths open to empirical disconfirmation? If someone told you that he was going to look for evidence that disconfirms 'There are round squares' would you consider him to be embarked upon a meaningful enterprise?

Don't answer these questions here. This discussion is off-topic in this thread.

[Edward Ockham]

Here is a tidied up version of my argument. Note the initial assumption.