Gary Ebbs published an extended version of his previous paper “Carnap on Ontology” in the JHAP a couple of years ago, but I’m still not convinced. The core of both papers is the claim that “Carnap’s method of identifying and eschewing ontological questions . . . stands or falls with his analytic-synthetic distinction. . .” (“Carnap on Ontology,” p. 54). The new paper now also seeks to defend Quine’s reading of “Empiricism, Semantics, and Ontology” (ESO), and I have to admit that Ebbs does a decent job of making Quine’s paper “On Carnap’s Views on Ontology” (OCV) seem surprisingly reasonable, if you’re willing to grant Quine’s priors. But I get lost when Ebbs positions this against what he calls the “new standard reading” of OCV, which conflates a number of quite disparate views from all over the map into a homogeneous phalanx of opposition to Quine. It’s possible that some of the named authors agree about some things, but Ebbs doesn’t characterize this “new standard view” precisely enough to get across what he’s really talking about, or what his criteria for inclusion were (he leaves a lot of anti-Quine things out of his list). I will stick with Graham Bird’s 1995 paper in Erkenntnis, for now, which Ebbs acknowledges as the earliest of those he positions himself against, and quotes most frequently. I will argue that Ebbs fails to address its main argument. If I’m right, then presumably whatever one may think about the “new standard view” more generally, Bird‘s critique, at least, of Quine’s arguments in OCV survives Ebbs’s cavils unscathed.

First, some preliminaries. Ebbs states unequivocally at the outset that “Quine was familiar with all of the logical, disciplinary, and philosophical assumptions that informed Carnap’s work.” (p. 1) But this assumption is clearly false. To repeat something I wrote here last year, in a different context:

Quine . . . persisted in regarding ontological questions as language-transcendent, and thought there was a “core meaning” to the question “what is there?” that persisted from ordinary (and traditional philosophical) language into his attempted explications of this question. For Carnap, it was of central importance that this made no sense; from sometime in 1924, ontological claims were self-consciously regarded as nothing but façons de parler; ontological questions were purely internal to a language framework. The “philosophical” final chapters of both the Aufbau and the Syntax are largely devoted to this priority; Carnap can’t be accused of not putting his cards on the table. But as Gary Kemp and Andrew Lugg point out, the question of what really ultimately exists was on Quine’s mind throughout his early career; they quote him saying: “In my youth I thought of the question of existence, or what there is, as perhaps the most basic question of philosophy and science. In the fullness of time the scales fell from my eyes.” They generously see his “youth” as stretching through Word and Object; certainly it includes the years when Quine claims to have been Carnap’s “disciple” — which raises the question how he could have missed this obvious, and pretty central, doctrinal clash between supposed master and supposed disciple.

So while I accept that Quine may well have read ESO from the peculiar perspective Ebbs goes to great lengths to reconstruct, there is no reason to think that Quine’s interpretation was therefore accurate or had any relevance to Carnap’s actual agenda. The Syntax that Quine read “as it emerged from Ina’s typewriter” in Prague during the summer of 1932 was the — pre-tolerance — first draft, not the final book in which tolerance is central. Quine never seems to have appreciated the fundamental role played in Carnap’s post-1932 development by the principle of tolerance. It is absent from Quine’s 1934 Harvard lectures on Carnap. The passages Ebbs cites from the Syntax to motivate Quine’s reading of ESO were (very likely) already in the first draft (we can’t be 100% certain because we have only the table of contents of that draft — which does however include section headings corresponding to §§76-7 of the published book, about “Allwörter,” on which Ebbs draws heavily).

On to Graham Bird, who in his 1995 paper — unlike Ebbs — was not engaged (or even much interested) in Quine exegesis. His main point was, rather, to argue that Carnap’s distinction between internal and external questions of existence does not presuppose or require an antecedent distinction between analytic and synthetic. So in fact Bird was not even really much interested in Carnap exegesis, in that paper, except insofar as the characterization of the internal-external distinction itself is in question. He doesn’t care whether Carnap actually uses an analytic-synthetic distinction in making his internal-external distinction; he wants to figure out whether Carnap needs to use it. His tentative answer is that Carnap doesn’t: “. . . so long as we can identify a specific language, in whatever way, we can also draw Carnap’s internal/external distinction. For once we have identified a language, then we can identify particular internal questions within it. . . It seems that all Carnap needs in order to draw his [internal-external] distinctions is the idea of an identifiable language, whether the items which identify it are to be further characterized meta-linguistically as analytic or not.” (pp. 56-7) This seems entirely consistent with Carnap’s own view that you could, if you were so inclined, construct a language without any analytic sentences in it at all; it would be a pretty useless language for just about all purposes languages are used for, but you could define such a thing. In such a language there would be no analytic-synthetic distinction. And there would still be a well-defined sense in which you could, relatively to such a language, classify any given sentence as belonging to that language or not, i.e. as internal or external to it. As Bird points out, Quine also appears to accept this basic idea of an identifiable language. Nor does Ebbs question it (or Quine’s acceptance of it); in fact he doesn’t address it at all.

What he does address are a number of issues unrelated to Bird’s main argument, but germane to Quine’s understanding of ESO according to Ebbs, e.g. the question whether Carnap’s examples of internal questions in both of his toy examples of linguistic frameworks — the “thing language” and the language of the natural numbers — are in both cases to be construed as analytic, or those in the “thing language” are rather to be regarded as empirical (i.e. synthetic). As far as I’m aware, just about every reader of ESO over the past 71 years — except one — has understood Carnap as meaning the latter. That exception, Ebbs argues, was Quine, who by dint of a portentous wedge driven between “analytic” and “trivially analytic” manages to make the toy examples of internal questions within the “thing language” come out analytic (I can’t reproduce Ebb’s tortuous reasoning here; it’s a long paper). I’m not sure whether I’m persuaded about Quine; Ebbs’s account, detailed as it is, seems more suggestive than conclusive. But whether or not I’m persuaded about Quine, I don’t see how Carnap’s own actual meaning can be mistaken for anything but the common understanding among all readers to date other than Quine. Ebbs cites Amie Thomasson as an example of this common (but in his opinion mistaken) reading, who in support of the common view quotes a typical passage from ESO (there are many similar ones): “The answers [to internal questions] may be found either by purely logical methods or by empirical methods, depending upon whether the framework is a logical or factual one. (ESO, 206; cited in Thomasson 2016, 123)” About which Ebbs has this to say:

Thomasson apparently reads this passage as saying, in effect, the following: the answers to internal questions in a logical framework, such as the framework for speaking about numbers, may only be found by purely logical methods; whereas the answers to internal questions in a factual framework, such as the framework for speaking about physical objects, may be found only by empirical methods. When read in this way the passage appears to support Bird’s [view]. (pp. 10-11)  

But this is a very far-fetched and tendentious way to interpret Thomasson’s (and everyone else’s) obvious interpretation of the ESO passage she quotes, which is not that the answers to internal questions in a factual framework may only be found by empirical methods (taking “empirical methods” not to include any mathematics or anything “analytic”), but rather that the answers to internal questions in a factual framework may also be found by empirical methods (or — equivalently — “only by empirical methods” if those are understood to include some mathematics, as in any scientific/factual framework since Galileo). So again, Ebbs’s argument against Thomasson’s understanding of her ESO passage fails to engage with what Thomasson (or Carnap) actually says.

Then there is the matter of Quine’s obfuscation of Carnap’s internal-external distinction by claiming that it derives from a more fundamental distinction between what he calls “category” and “subclass” questions. Ebbs not only endorses Quine’s diagnosis (and makes a plausible case for the motivation behind it, in Quine’s mind), but, in the course of his paper, drifts into downright attribution of Quine’s distinction to Carnap, referring e.g. to “Carnap’s use of the category-subclass distinction to critique philosophical questions of existence” (p. 13). Carnap, however, not only did not use that distinction; he rejected it explicitly. In his notes for the colloquium talk in Chicago where Quine presented CVO, he says (like a number of later commentators) that Quine’s category and subclass questions are both internal. And that while he sympathizes to some degree with Quine’s program of eliminating (or ignoring) external questions altogether, that also undermines the possibility of criticizing (or entering into any sort of dialogue) with philosophers who continue to ask, and attempt to answer, external questions. Shouldn’t one try to make clear what is wrong with that?

I wish Ebbs had limited the scope of this paper to an interpretation of Quine’s (mis)understanding of ESO, and stopped there. Where it strays beyond that, as the examples above illustrate, it tends to fall back into the bad old species of ill-informed “Quine right, Carnap wrong” dogmatism that Ebbs has himself over the years made substantial contributions toward overcoming.