The Road to Ramseyan Humility

I’ve mentioned to a few of you around here that I think it is Lewis’ solution to “Putnam’s Paradox” that lands him in “Ramseyan Humility”. Here’s a sketch of why. I apologize for assuming some background familiarity with both “Putnam’s Paradox” and “Ramseyan Humility”, but I’m trying to keep things punchy.

Consider Lewis’ argument for Humility. Lewis’ assumes that T is realized by fundamental properties, which (because fundamental properties are plausibly wholly distinct from the roles they realize) allows Lewis to run the permutation argument to establish multiple possible realizations of T. But why does Lewis assume that T is realized by fundamental properties? There are actually several questions here. First, why does Lewis think that T is even prima face the sort of thing that is realized—that is, why does Lewis think that the Ramsey sentence approach to the content of T is the right one? Second, why does Lewis think that T is realized (rather than holding a non-representational interpretation of the Ramsey sentence as Sklar does)? Third, assuming that T is realized by something, why does Lewis think that T is realized by (metaphysically robust) natural properties [rather than, say, (metaphysically thin) classes]? And, finally, fourth, assuming that T is realized by natural properties, why does Lewis think that T is realized by completely natural properties, i.e., fundamental properties (rather than, say, natural “enough” properties).

Although there are four questions here, I think Lewis’ answer to the first three will be roughly the same: realism. His answer to the fourth question, however, can be found in RH where he makes a particular observation about scientific progress. Let’s take these questions in order.

Why does Lewis think that T is even prima face the sort of thing that is realized—that is, why does Lewis think that the Ramsey sentence approach to the content of T is the right one? Presumably Lewis thinks that the RS approach to the content of T is the right one because he thinks that it is the most plausible realist theory about the reference of terms that are first introduced into a language as part of a new theory (that is, t-terms). Presumably this theory wins out over its competitors by avoiding their implausible implications and not generating any of its own.

Why does Lewis think that T is realized (rather than holding a non-representational interpretation of the Ramsey sentence)? To hold a non-representational understanding of the Ramsey sentence for a given theory is to understand the Ramsey sentence as simply a way of “book keeping” in our theorizing. On this view, the role of the Ramsey sentence is to simplify a theory, the actual content of which is given by its empirical consequences alone, by making those consequences all derivable from some more manageable sentence (i.e., the Ramsey sentence). But the proponent of such a view, by the realist’s lights, is trying to reap the benefits of manageability without the metaphysical costs—that is, without taking himself to be committed to there actually being an x such that… This, again by the realists’ lights, is an illegitimate more, for it would be remarkable fact indeed if the theory be both successful and derivable from the Ramsey sentence without the Ramsey sentence actually being (approximately) true (that is, true in the representationalist sense).

Why does Lewis think that T is realized by (metaphysically robust) natural properties [rather than, say, (metaphysically thin) classes]? In an attempt to block Putnam’s “model-theoretic argument” (what Lewis calls “Putnam’s paradox”), Lewis adopts the view that an intended (that is, correct) interpretation of a language is one that does the best on balance of both making the total body of theory in that language come out true and assigning maximally natural properties as the referents of the terms. This is Lewis’ means of blocking Putnam’s anti-realist conclusion that “almost any consistent theory is true of almost any world”. Thus, the terms of T, on Lewis’ view, refer to (at least somwhat) natural properties, and since some of those terms are the t-terms (which were existentially generalized on to form the Ramsey sentence), it follows that T is realized by (at least somewhat) natural properties.

Finally, why does Lewis think that T is realized by completely natural properties, i.e., fundamental properties (rather than, say, natural “enough” properties)? In RH, Lewis writes

Scientific theorizing and the discovery of fundamental properties have gone hand in hand… If T is the limit (perhaps never reached) of a process in which theorizing and the discovery of fundamental properties go hand in hand, then the fundamental properties mentioned in T will be named by T-terms. (3)

I’ll leave this without comment for now (since I have no idea what to say about it).

If I’m right in the above, then Lewis’ solution to Putnam’s paradox alone does not land him in Ramseyan Humilty. However, it does play a crucial role in getting him there. The question I’d like to explore next is whether the causal-theory-of-reference solution to Putnam’s paradox also threatens to land the causal theorist in (an analog of) Ramseyan Humility. I think that prima facie it doesn’t. But I’d like to know if this is merely prima facie. I think it may be.

By the way, here’s an interview with Susan Haack about her 2003 book “Defending Science Within Reason” (click “listen now” on the right hand side).

Advertisements

2 Responses to The Road to Ramseyan Humility

  1. Alex S. says:

    Hey Dustin,

    Interesting project; keep us posted on how it turns out. Just a point of clarification. You mention above that

    “Lewis thinks that T is realized by completely natural properties, i.e., fundamental properties (rather than, say, natural ‘enough’ properties).”

    There are at least two types of property that Lewis could be thinking of as “fundamental” here: (1) properties that depend for their instantiation upon no other properties, yet are that upon which all other properties depend (presumably, these are the properties of microphysics), or (2) properties that are perfectly natural in the Lewisian sense of “perfectly natural” (these are the properties that figure into lawlike generalizations and precise comparisons of qualitative similarity, and which include but are not necessarily exhausted by the properties of microphysics).

    Thus my question: in this context are you (and Lewis, for that matter) speaking of properties that are 2-fundamental, or 1-fundamental? (Maybe nothing turns on this, but the answer would help me begin to see how Lewis might have answered your third and fourth questions.)

  2. dtlocke says:

    Hi Alex,

    By “2-fundamental property” do you just mean “property that figures into lawlike generalizations and precise comparisons of qualitative similarity”? And by “1-fundamental property” do you just mean “property that depends for its instantiation upon no other properties, yet upon which other properties depend”? If so, then the answer to your question is definitely 1-fundamental.

    However, the answer might ALSO be 2-fundamental. We should tease apart the two aspects of your 2-fundamental. Let’s say that the 2-fundamental properties are the ones that “figures into lawlike generalizations” (a contingent matter for Lewis) and let’s say that the 3-fundamental properties are the ones that “figure in precise comparisons of qualitative similarity” (a necessary matter for Lewis, I think). In that case, Lewis believes that some but not all of the 1-fundamental properties are 2-fundamental properties, some but not all of the 2-fundamental properties are 1-fundamental properties, and all and only the 1-fundamental properties are 3-fundamental properties.

    The reason Lewis would say that not all of the 1-fundamental properties are 2-fundamental is that he thinks there are “idler” and “alien” 1-fundamental properties–that is, properties that are instantiated in the actual world but play no role in the workings of nature and properties that are not instantiated in the actual world, respectively. The reason Lewis thinks that (probably) not all of the 2-fundamental properties are 1-fundamental is that he thinks that the laws are the theorems of the axiom system that does best on balance of power and simplicity, and achieving the best balance may require that some of the laws describe the workings of non-1-fundamental properties. Finally, the reason Lewis thinks that all and only the 1-fundamental properties are 3-fundamental is that he believes that the two are analytically equivalent (I think, but see “New Work for a Theory of Universals” for confirmation, or disconfirmation).

    Thanks for the comment!

%d bloggers like this: