NATHANIEL GOLDBERG | Kantian Conceptual Geography | Oxford University Press 2015


By Nathaniel Goldberg 

I am grateful to Paul Franco for his penetrating critique. As Franco knows, Kantian Conceptual Geography (hereafter ‘KCG’), though not quite a work in the history of philosophy, nevertheless takes that history seriously. Whether or not I hit the right mix in writing, this kind of work is probably not the easiest for reviewing. And yet Franco was able to offer thought-provoking comments. Hopefully my reply does them justice.

KCG explores issues in analytic epistemology, philosophy of language, and metaphysics by appealing to theses drawn from Kant’s Critique of Pure Reason:

Dualism: All empirical concepts, terms, or properties are linked essentially to subjective and objective sources.

Principlism: The subjective source of all empirical concepts, terms or properties takes the form of subjective principles.

Kantianism combines these:

Kantianism: All empirical concepts, terms or properties are linked essentially to subjective and objective sources. The subjective source may take the form of subjective principles.

Though all three theses have ontological and epistemological variants, each ‘subjective source’ means the specific conceptual, linguistic or perceptual capacities in an individual subject’s mind, shared across a group of subjects’ minds, or encoded in a subject’s or subjects’ language or (other) conventions. ‘Objective’ means not subjective.

My focus in KCG was on exploring those epistemological, semantic and metaphysical issues—including analyticity, apriority, empirical properties, incommensurability, meaning, possible worlds, response-dependence and truth. Nevertheless I also wanted to present defensible readings of Kant to do so. Understandably Franco brackets my discussions of many of the issues just mentioned as well as philosophers articulating them. Instead Franco has questions about three areas central my project:

  1. I’d like to hear more about the senses of “conceptual, linguistic, or perceptual capacities” that render, say, the following all subjective in relevantly similar senses: a transcendental mind, genetic mechanisms shaped by (objective?) processes of evolution, linguistic conventions, and agreement/consensus among a scientific community.

  2. I would like to hear more from Goldberg about the ways in which he thinks Principlism optional to Kantianism.

  3. I’d be interested, then, in hearing from Goldberg about the ways in which he thinks his form of conceptual geography is similar to and different from the very brief and rough characterisation of historical epistemology that I have given here.

Franco is being unnecessarily kind. He would like to hear more from me because, whether he explicitly intends to do so or not, I read him as expressing three worries associated with the first two areas, and I see three responses to his final area. I address each area—and ensuing threesome—in turn.

1. Subjective Sources

I read Franco as expressing three worries about my use of ‘subjective’ in particular.

First, some constitutive principles (those are his examples; as Franco explains, on my view constitutive principles are the ontological version of subjective principles) that I claim are subjective are not. Franco appeals to work of Hans Reichenbach and Michael Friedman, two philosophers whom I classify as Kantian. (Reichenbach likely would and Friedman definitely does classify himself as such also.) Nonetheless, according to Franco, Reichenbach and Friedman do not understand constitutive principles as subjective in my sense. “For both Reichenbach and Friedman”, Franco writes,

Kant’s brand of transcendental inquiry is better framed as aimed at uncovering the conditions of certain secure and dependable bodies of knowledge in a way that avoids references to a transcendental mind; as such, I take Reichenbach and Friedman as reluctant to appeal to things like conceptual, linguistic or perceptual capacities.

Franco knows Reichenbach’s view better than I do. Regardless, as Franco and I agree, Reichenbach wanted to retain much of the structure that Kant thought knowledge had, distinguishing a priori from a posteriori or empirical elements. But how could Reichenbach do so without appealing to transcendental idealism? I argued that Moritz Schlick persuaded Reichenbach that he “should designate [his] a priori principles as conventions” (Schlick 1979:324) in Henri Poincaré’s fashion (KCG, p. 148). Although I am unsure whether Reichenbach explored the nature of these conventions, I am sure that his (and Poincaré’s own) conventionalism inspired Rudolf Carnap to take conventions to be linguistic. And linguistic conventions establish which linguistic capacities are appropriate for users to have insofar as they are users of that language and so members of that linguistic community. This does not mean that Reichenbach would endorse my notion of subjectivity as concerning linguistic capacities, but it does mean that Reichenbach should be less reluctant to appeal to such capacities than Franco thinks.

Franco next observes that Friedman elaborates on Reichenbach’s analysis by describing discrete elements of physical theories: (a) “completely abstract mathematical equations” (Franco) and (b) “a special class of non-empirical physical principles […] whose function is precisely to establish and secure the required connection between abstract mathematical structures and concrete sensory experience” (Friedman 2001:78). Though Franco does not mention this, Friedman also describes a final element: (c) properly empirical laws directly tested against experience (2001:35, 38, 45, 74, 78, 79–80, 83) (KCG, Chap. 6, §5). Friedman in particular thinks that, while (a) and (b) are a priori and non-empirical, (c) is empirical. Franco worries specifically about how (b) can count as subjective in my sense. Though non-empirical, it is nevertheless physical. In Reichenbach’s and Friedman’s shared term, it “coordinates” between purely abstract mathematics and sensory experience in a way that allows properly empirical laws to be formulated and tested against experience.

Franco illustrates his worry with what Friedman (and, on both their readings, Reichenbach) takes to be an example of (b), the light principle of special relativity, according to which the speed of light in a vacuum is constant for all observers. Franco claims that on my view the source of the light principle seems objective since it is derived from the nature of light itself. Though Franco does not put it thus, because by my definition of ‘objective’ it is not subjective, I am caught in a contradiction.

Yet, according to Friedman, the light principle does not function in special relativity as an empirical claim. As Franco himself observes, according to Friedman, “Einstein has ‘elevated’ an empirical law to the status of a convention” (2001:88). Perhaps I am weighting this move more than Franco is. But that is because Friedman weights it more too. According to Friedman, for Einstein, the light principle becomes a relativised a priori constitutive principle in the sense of involving “an essentially non-empirical element of ‘decision’” (2001:88) in its use. And that decision is based on the capacities of subjects—including linguistic, conceptual and perceptional—to formulate and use it to coordinate between abstract mathematics and sensory experience.

Friedman also says that “Einstein uses his light principle empirically to define a fundamentally new notion of simultaneity and, as a consequence, fundamentally new metrical structures for both space and time” (2001:88; emphasis his). Friedman’s italics are telling. The light principle defines, and so stipulates—or establishes by convention—a new notion and new structures. Because definitions are linguistic, their resulting convention would be linguistic too. The light principle does this empirically because this linguistic convention is meant to apply to experience. The light principle is subjective yet establishes the legitimate terms with which empirical laws can be formulated and tested against experience. While Franco “find[s] it odd to call Friedman’s physical, but non-empirical coordinative principles subjective in the sense of being traceable back to our specific conceptual, perceptual or linguistic capacities”, on Friedman’s own view they seem to be just that. While Franco does grant that conventions “are non-empirical in the sense of not being fully determined by experience” he nevertheless continues,

it does not seem to me to be the case that ‘non-empirical’ (or ‘not determined by experience’) and ‘subjective’ (in Goldberg’s sense of being traceable back to specific linguistic or conceptual capacities) are co-extensive, especially in the case of the principles Friedman calls mathematical-physical presuppositions of our empirical knowledge.

Perhaps they are not co-intensive.[1] But I do think that they are co-extensive.

There is more evidence for my reading of Friedman. Besides championing Reichenbach, in characterising his a priori constitutive principles Friedman (2001) is also inspired by Carnap and Thomas Kuhn. As I quote in KCG, Friedman urges that “one could attempt to combine basic aspects of Carnap’s philosophy of formal languages or frameworks with fundamental features of Thomas Kuhn’s much less formal theory of scientific revolutions” (2001:xii). Doing so, Friedman continues, allows one to avoid “the drawbacks of Carnapian formal Wissenschaftslogik. In particular, W.V.O Quine’s well-known and widely accepted attack on the Carnapian conception of analytic truth need no longer compel us to adopt a thoroughgoing epistemological holism according to which there is nothing left of the a priori at all”.[2]

There are two ways to read what Friedman thinks Kuhn brings to his arranged marriage with Carnap. Kuhn brings a way in which Carnap’s conception of analytic truth can be made either (a) informally linguistic or (b) non-linguistic. Since Friedman claims that Carnap’s reliance on “formal languages” and “Wissenschaftslogik” (or logic of science) makes his conception of analytic truth vulnerable to Quine’s attack, either seems on Friedman’s view to save Carnap’s conception. Now, if Friedman tries to save Carnap’s conception by reformulating it as informal linguistic, then he is appealing to informal linguistic conventions. And linguistic conventions still establish linguistic capacities. Conversely, if Friedman tries to save Carnap’s conception by reformulating it as non-linguistic—if what he himself calls “relativized a priori constitutive principles” are themselves non-linguistic—then two things follow. First, Friedman owes us an explanation of what they are. For it is unclear what a non-linguistic principle that nevertheless is used in generating empirical claims (which are linguistic if anything is) could be. Since Friedman gives no explanation, we should presume that he retains a role for language. Hence he retains a role for linguistic capacities. Second, were his constitutive principles somehow non-linguistic, then they would still be conceptual. Rather than being empirical, they would establish the legitimate concepts with which a properly empirical law could be formulated and tested against experience. But then they would establish conceptual capacities for such formulating and testing. Either way, Friedman’s constitutive principles are traceable back to specific linguistic or conceptual capacities.

Moreover, as I argue in KCG (Chap. 5, §6), Friedman has no reason not to regard his relativised a priori constitutive principles as linguistic. There, I identify six arguments that Quine makes against Carnap’s conception of analytic truth[3] and show that none impugns Friedman’s relativised a priori constitutive principles. Rather, in each case, their historical rather than any informal linguistic or non-linguistic nature saves them. So nothing is preventing Friedman from treating his principles as linguistic, informal or otherwise. Since, I have argued, Friedman should treat them as linguistic or conceptual, he should treat them as traceable to linguistic or conceptual capacities. That makes them subjective in my sense.

The second worry that I read Franco as expressing about my use of ‘subjective’ is that its referents are a gerrymandered set. Franco asks:

What unifies [1] non-empirical and non-physical principles (like geometrical conventions), [2] physical principles that function non-empirically within empirical theories (like Einstein’s light principle), [3] methodological principles (like Davidson’s principle of charity), and [4] principles like Kant’s cause and effect that are explicitly traced back to a transcendental mind? (bracketed numbers mine)

To these let me add those that at the outset I quoted Franco as listing:

[5] a transcendental mind, [6] genetic mechanisms shaped by (objective?) processes of evolution, [7] linguistic conventions, and [8] agreement/consensus among a scientific community. (bracketed numbers mine)

Franco is right that I call [1]–[8] ‘subjective’. Is anything non-arbitrary holding the set together? Though the list is long, it is worth considering in toto:

[1] seems least objectionably subjective. Geometric conventions, Franco’s example, have since Poincaré been seen as stipulations or—and I think these are co-extensive—elements of a formal language. And stipulation and formal-linguistic elements establish the legitimate terms with which empirical laws can be formulated and tested against experience. They therefore establish linguistic capacities for those who accept and use them. Moreover above I explained that this is the sense in which [2] is subjective. For Friedman, Franco’s own example should rely on linguistic or conceptual capacities.

[3] requires more discussion. While Franco has my analysis of Davidson in mind, and while Davidson maintains that the principle of charity derives from the subject’s “underlying methodology of interpretation” (2001b:197), that principle is not merely methodological. It is instead, as Davidson (2001a:220–1) himself maintains, constitutive. Davidson even calls the principle “synthetic a priori” (2001a:221). Moreover, as Kant himself holds that “[w]e cannot think any object except through categories” (B165), embedded in his synthetic a priori judgements (or principles), so Davidson holds that we cannot interpret any utterances except through the principle of charity. Perhaps I should have quoted Davidson here: “[C]harity is not an option, but a condition of having a workable theory [of interpretation] […]. Charity is forced on us; whether we like it or not, if we want to understand others, we must count them right in most matters” (2001b:197). The connection between Davidson and Kant is closer than even Davidson realises. One conclusion of KCG is that, while Davidson famously attacks Dualism, he is instead committed to it (Chapter 5). Likewise, in the form of the principle of charity, he is committed to Principlism too (Chapter 4). Because Franco does otherwise bracket Davidson, I do not mean any of this as a criticism of Franco’s view but instead as an elaboration of Davidson’s and by extension mine. Regardless, in KCG I do not consider any methodological principles. So I need not explain the sense in which [3] is subjective.

In fact Davidson’s principle of charity has much in common with [4], which traces to [5], and Franco himself seems not to question the subjectivity of either. His example of [4] is Kant’s Second Analogy, which presupposes the category of causality. Of the Transcendental Deduction of the categories Kant himself observes: “Thus a difficulty is revealed here that we did not encounter in the field of sensibility, namely how subjective conditions of thinking should have objective validity” (A89–90/B122). The categories, and the Analogies and other principles that presuppose them, are subjective in Kant’s own sense, which is subsumed by mine. The categories are conceptual capacities if anything is, and they are encoded in a subject’s mind. I return to this below.

[6] requires more discussion too, since I answer Franco’s parenthetical question affirmatively. Genetic mechanisms are shaped by processes of evolution—even though those mechanisms are subjective, while those processes are objective, in my sense. Here my explanation concerning [2] applies. Just as Einstein consciously elevated an empirical law to the status of a convention, subjects subconsciously elevate the empirical result of evolution—their genetic mechanisms—to the status of innate dispositions and therefore broadly conceptual, linguistic, and perceptual capacities. While the light principle understood before Einstein might have been formulated and tested against experience, Einstein stipulated its truth to formulate and test his own properly empirical laws as such. While genetic mechanisms are shaped by objective processes of evolution, subjects use those mechanisms in the form of broadly conceptual, linguistic, and perceptual capacities when they formulate and test their own hypotheses against experience. I return to this below too.

As for [7] and [8], a community’s linguistic conventions codify its agreement/consensus. And those conventions are subjective in the sense in which [1] and [2], and broadly [6], are. Non-empirical and non-physical principles, like geometrical conventions, are linguistic conventions, which are agreements/consensuses among the scientific community. Physical principles that function non-empirically, like Einstein’s light principle, are, for reasons already discussed, linguistic conventions that are agreements/consensuses among the scientific community also.

Is the set gerrymandered? [1], [2], [6], [7], and [8] all concern linguistic capacities. While the linguistic capacities involved in [6] are broader than those involved in the others, that is the point of my distinguishing subjective scopes. [6] might ground anthropocentric linguistic capacities, while [1], [2], [7], and [8] might ground ethnocentric ones. As Franco observes, I identify these and other subjective scopes too. What should we make of the remaining elements? [3] may be either set aside or, if understood as transcendental, elided with [4] and [5]. Now, as I observed above, Franco himself writes: “For both Reichenbach and Friedman, Kant’s brand of transcendental inquiry is better framed as aimed at uncovering the conditions of certain secure and dependable bodies of knowledge in a way that avoids references to a transcendental mind”, and, as I showed, both Reichenbach and Friedman can understand those conditions as concerning linguistic capacities. More generally, as Alberto Coffa (2008) argues, the “semantic tradition from Kant to Carnap”, in which Reichenbach himself was central, was concerned with understanding Kant’s claims about intuition and other transcendental faculties in terms of semantic, and so linguistic, convention (KCG, pp. 4n.1, 69n.15, 144n.12, 145n.14, 148n.17). Insofar as [4] and [5] seem dissimilar from [1], [2], [6], [7] and [8], that is because the latter are a semantic version of the former. But that is to say that linguistic capacities are a semantic version of conceptual capacities, which is my point. Hence, even given the variety of its elements, [1]–[8] is not a gerrymandered set.

Admittedly that does not mean the subjective/objective distinction is the most perspicuous way to characterise Kantianism or Dualism specifically. That is the third worry that I read Franco as expressing about my use of ‘subjective’. He asks:

Why think the subjective/objective distinction is the relevant one for charting Kantian territory and uniting the disparate positions Goldberg canvasses, instead of, say, other Kantian distinctions that may (or may not) map directly onto the subjective/objective distinction like the [a] constitutive/empirical distinction, the [b] form/content distinction, or the [c] passivity (sensation)/spontaneity (convention) distinction? (bracketed lettering mine)

In KCG (Chap. 5) I explain that the subjective/objective distinction just is Davidson’s scheme/content distinction, which is a form of [b], and that Davidson’s dualism is my Dualism of subjective and objective sources. So Franco’s [b] is my distinction. Though [c] may be vaguer than the subjective/objective distinction, these may amount to the same also.

There are two reasons, however, that the subjective/objective distinction differs from [a] and so that I cannot accept it. On the one hand, as Franco observes, I regard Philip Pettit as concerned not with constitution but with acquisition. For him, all empirical concepts, terms or properties are acquired by a subject’s appealing essentially to subjective and objective sources. Moreover, as he also observes, I read Henry Allison (2004) as construing Kant’s own transcendental idealism similarly. The subjective/objective distinction has to do with more than merely constitution. [a] falls shorts.

On the other hand, for all Kantians, the empirical derives from the interanimation of the subjective and the objective. For Kant, empirical judgements, and experience itself (B147), result from sensibility’s intuiting sensation in space and time, whose resulting empirical intuitions the understanding conceptualises with the categories. For Carnap, synthetic statements, which are empirical and so describe experience, result from analytic statements being used to formulate them given sense data (2003) or sensation (1988). For Kuhn, empirical claims result from the application of paradigms (2012) or lexica (2002) to sensation. For Reichenbach as Friedman (2001) reads him, and Friedman himself, properly empirical laws are formulated by using physical non-empirical principles to coordinate purely abstract mathematics and sensory experience—which in this sense just means sensation. [a] therefore presupposes the subjective/objective distinction.

Hence my distinction just is [b], may amount to [c], and is superior to [a].[4]

2. Requiring Principlism

Franco’s concern with what I mean by ‘subjective’ targets the required Dualism component of Kantianism. Franco would also like to hear more about why I think that the Principlism component is itself merely optional. I read him as offering three worries that it should be required too.

First, making Principlism required solves what Franco takes to be my dubious uses of ‘subjective’. His point seems to be that since we better understand when a principle is, rather than what makes it, subjective, I might make the existence of subjective principles necessary for any view to be Kantian. Insofar as I have shown that my uses of ‘subjective’ are not dubious, however, this first worry is avoidable.

Second, Principlism is essential to various Kantian positions. Indeed, as I observe, Kant’s principles, his synthetic a priori judgements, are “indispensable” (KCG, p. 7) to his view. So are Kuhn’s principles implicit in his paradigms, disciplinary matrices and lexica. So is Davidson’s principle of charity. And so are Carnap’s analytic truths, and Reichenbach’s and Friedman’s relativised a priori constitutive principles. Nevertheless, just because Principlism is essential or indispensable to various Kantian positions does not mean that Principlism is or should be essential or indispensable to Kantianism. Nor, independent of other reasons to do so, need it encourage me to make it as such. One independent reason that Franco offers was the one just considered, that making Principlism required would help with my uses of ‘subjective’, though these turn out not to be dubious. Another is that leaving Principlism optional makes other views that we have reasons not to consider Kantian, like Quine’s holism and Pettit’s global response-dependence, Kantian.

This is the third worry that I read Franco as offering for making Principlism required. Franco questions Quine’s Kantian credentials in particular. Friedman is right that I make much of Quine’s claim that “[t]aken collectively, science has its double dependence upon language and experience; but this duality is not significantly traceable into the statements of science taken one by one” (2006:42)—and that I understand language, as depending on linguistic capacities, as subjective, while experience here is objective. Franco (note 8) is also right to question whether Quine’s use of ‘experience’ does not succumb to my own point about the empirical, viz., that it is hybrid. Anticipating that worry I mentioned (KCG, p. 113n.4) that Quine also describes subjective and objective sources in terms of “language” and “extra-linguistic fact” (2006:34). Nevertheless, Franco is concerned that, when Quine goes on to claim that he “espouse[s] a more thoroughgoing pragmatism” than Carnap and C.I. Lewis (Quine 2006:46), he might be denying the usefulness of Dualism altogether. Because Quine rejects Principlism—that is how I read his rejecting analyticity (KCG, Chap. 6, §3)—leaving Principlism optional apparently dilutes Quine’s commitment to Dualism so much that he turns out not to be a Kantian in any useful sense.

Quine’s commitment to Dualism is nevertheless not diluted. Though I do not quote this in KCG, Quine does write:

We cannot strip away the conceptual trappings sentence by sentence and leave a description of the objective world; but we can investigate the world, and man as a part of it, and thus find out what cues he could have of what goes on around him. Subtracting his cues from his world view, we get man’s net contribution as the difference. This difference marks the extent of man’s conceptual sovereignty—the domain within which he can revise theory while saving the data. (1964:5)

Man’s (and woman’s) net contribution, our “conceptual trappings” (traceable to conceptual capacities), is the subjective source of sentences and therefore terms. The “data”, which just are of the “objective world”, are the objective source. Even the title of Quine’s final book, From Stimulus to Science (1995), names the objective source of empirical concepts and terms and then their systematisation into science, which, like the empirical concepts and terms themselves, are hybrid.

There is another reason to count Quine a Kantian. Quine maintains:

The lore of our fathers is a fabric of sentences […]. It is a pale gray lore, black with fact and white with convention. But I have found no substantial reason for concluding that there are any quite black threads in it, or any white ones. (2004:87)

The lore of our fathers (and mothers) has conventional and factual sources. The conventional source is subjective; the factual, objective. Dualism is true. Nevertheless, there are no quite white threads. No knowledge claims are purely subjective. There are no subjective principles. Principlism is false. Yet neither are there any quite black threads. No knowledge claims are purely objective. The world’s intrinsic nature or aspect, independent of our subjective responses to it, is unknowable.

In KCG I follow Pettit (2002, overview to part I, essay 3) in calling this latter view ‘noumenalism’ and show that all Kantians endorse it. Quine endorses it here, as he does above when he admits that “[w]e cannot strip away the conceptual trappings sentence by sentence and leave a description of the objective world”—a world objective in my sense. Kant (in)famously endorses it, and I discuss (KCG, Chap. 2, §3) different construals of his particular variety. I also discuss (KCG, Chap. 2, §§3–5) how Pettit endorses noumenalism and engages in a debate with Michael Smith and Daniel Stoljar over how to construe it. As I explain (KCG, pp. 74, 204), Kuhn endorses it too: “Underlying all these processes of differentiation and change, there must, of course, be something permanent, fixed, and stable. But, like Kant’s Ding an sich, it is ineffable, undescribable, undiscussable” (2002:104). As I argue (KCG, p. 134), Davidson endorses it as well. And I argue as well (KCG, p. 149n.20) that Carnap in the form of non-conceptual sense data (2003) and non-conceptual sensation (1988) endorses a version of noumenalism too—with similar reasoning applying to Reichenbach and Friedman.

In fact, as I show, noumenalism follows directly from Dualism (KCG, p. 56). By requiring that there be subjective and objective sources, Dualism requires that there be an objective one—a source independent of the subject. That source might be the world in itself, if noumenalism is construed ontologically, or the world considered in itself, if noumenalism is construed epistemologically. Regardless of how noumenalism is construed, however, all empirical concepts, terms or properties would have subjective and objective sources mixed. But then possessing such concepts or terms, or perceiving such properties, would not allow us to know any objective source in itself or considered in itself. Conversely, any non-empirical concepts, terms or properties (were there any) would be purely subjective. So possessing such concepts or terms, or perceiving (if this is even the right word) such properties, would not allow us to know any objective source in itself or considered in itself either. And those are the only options for the Kantian. We can then never know any objective source in itself or considered in itself.

No matter how we understand noumenalism, every kind of Kantianism entails it in virtue of its Dualism component. Because noumenalism has historically and contemporarily been associated with Kantianism, we therefore have reason to count anyone committed to Dualism—even if, like Quine and Pettit, one is not also committed to Principlism—as a Kantian.[5]

3. Conceptual Geography vs. Historical Epistemology

Finally Franco would like to hear about “the ways in which [Goldberg] thinks his form of conceptual geography is similar to and different from […] historical epistemology.”

As I understand it, historical epistemology is related to Michel Foucault’s (2007) method of genealogy. Historical epistemology aims to understand the historical (or ‘genetic’) origin of our concepts, thereby illuminating their contemporary content as well as that content’s historical contingency.[6] This suggests three differences from conceptual geography.

First, while historical epistemology may be concerned with conceptual geography,[7] it must be concerned with conceptual genealogy. A historical epistemologist’s investigating the (average? dominant?) contemporary concept of substance would recognise that we inherited it from various concepts employed by the Early Moderns, who inherited theirs from various concepts employed by the Scholastics, who inherited theirs from Aristotle (who inherited his from Plato, tracing earlier still, much earlier being historically inaccessible). Though the concept of substance has changed—conceptual inheritances, like monetary ones, can be depleted, replenished, and diversified—we nevertheless inherited some of its virtues and vices. And understanding all that helps us understand what our concept is and that it need not have been as such. Though both are worthwhile, conceptual geography and conceptual genealogy are nevertheless distinct, and historical epistemology might forgo the former.

Second, conversely, while conceptual geography may be concerned with conceptual genealogy, it must (obviously) be concerned with conceptual geography. By ‘engaging in conceptual geography’ I mean exploring, surveying, and mapping how concepts relate to one another and the broader conceptual world (KCG, p. 4). As Franco observes, I get my notion of conceptual geography from Gilbert Ryle’s (2000:7) notion of logical geography. Ryle does not engage in genealogy, nor need he. Nor need any conceptual geographer. While it is possible to combine them, conceptual geographers can accept their concepts at face value without researching their inheritance.

And third, because only historical epistemology must be concerned with genealogy, only it must be concerned with history. That concern must also be deep. To understand our concept of substance, a historical epistemologist should consider not just major but also minor Early Modern works, including those lost to most contemporary readers. She should also consider religious, scientific, and other texts from the era. She should dig into historical ground, not for history’s own sake, but for the sake of the present.

Now, though conceptual geography need not be concerned with conceptual genealogy or history (deep or otherwise), it can be. Indeed my own brand of conceptual geography is concerned with both. Although I do not provide a genealogy of most of my concepts, I do provide something of one concerning the definition of ‘Principlism’ and ‘analyticity’, since I defend Principlism partly by “consider[ing] the historical progression of versions of Principlism” (p. 137) and arguments against it, which involves considering the historical evolution of analyticity (KCG, Chap. 6). Moreover, though I do not engage with history deeply, I do appeal to historical views from major historical works. But should I do more?

As with all conceptual geographers, the answer is that I could but need not. There is however an overall historicist lesson that I should make clear. No philosophical method is ahistorical. Conceptual geography, no matter what brand, is necessarily situated in the historical period in which it is conducted. My own conceptual geography, which appeals to historical authors to make a present point, does so in a present moment in history that is itself influenced by past moments. Even my idea of engaging in conceptual geography occurs in such a moment. Not enough appreciate William Faulkner’s insight that “[t]he past is never dead. It’s not even past” (1994, I.iii), to which I should add that the present and the past are both historical. Implicitly holding to Faulkner’s insight, I made sure never to claim that my definitions of ‘subjective’ and ‘objective’, or ‘Principlism and Kantianism’, or my engaging in Kantian conceptual geography generally, are given or performed sub specie aeternatis. But that is distinct from explicitly declaring that they are not. I might have even observed that a different Kantian conceptual geography could be performed in every generation.

To be sure, Franco does not take historical epistemology to be superior to conceptual geography, my brand or generally. Instead he explains:

Just like Goldberg’s conceptual geography does, historical epistemology might yield somewhat surprising results and help us understand the conceptual space current philosophers operate in. For example, Scott Edgar argues in his examination of neo-Kantian conceptions of objectivity in the 1850s to the 1870s that “neo-Kantians [e.g., Hermann von Helmholtz, F.A. Lange, and Otto Liebmann] […] reject the view that our knowledge’s objectivity consists in resembling or being determined by mind-independent objects.” (2015:103; Franco’s ellipsis)

Although I do not know Edgar’s work, and though I quote Liebmann, I am no expert on these early neo-Kantians, I do discuss the two dominant neo-Kantian schools, the Marburg (including Hermann Cohen, Paul Natorp, and Ernst Cassirer) and the Baden or Southwest (Wilhelm Windelband, Heinrich Rickert, Jonas Cohn, Emil Lask, and Bruno Bauch) on a related topic (p. 22n.28). Nevertheless, I accept Franco’s general point that historical epistemology might itself yield surprising results concerning the conceptual space in which current philosophers operate.

For that reason, I am, like Franco, a methodological pluralist. On my view, historical epistemology has its place, as does my brand of conceptual geography, as do other brands and other methodologies altogether. Toward the beginning of KCG (p. 4) I quote Wilfrid Sellars, who himself appeals to the history of philosophy, and especially Kant, to make progress on “the current [philosophical] situation and its demands” (1968:1). At the end of my response to Franco’s stimulating critique of KCG let me quote Sellars again: “The aim of philosophy, abstractly formulated, is to understand how things in the broadest possible sense of the term hang together in the broadest possible sense of the term” (1991:1). One way to assist in achieving this aim is for philosophers to embrace the broadest possible philosophical tools—historical epistemology and conceptual geography among them.

Acknowledgements: Thanks go to Chris Gavaler for helpful comments.

Received: 22 August 2016.


1. Franco (note 12) explains that the same point apparently holds for Kuhn’s paradigms and Carnap’s linguistic frameworks. But, even if it does hold for Kuhn’s paradigms—and below I offer reasons to think that it does not—Kuhn’s replaces his paradigms with lexica, which are linguistic, and Carnap’s frameworks already are. And linguistic objects establish linguistic capacities. Moreover Franco seems to elide two separate points. It is one thing to say, as I am saying, that coordinating principles, linguistic frameworks, and lexica are subjective. It is another to say, as I am not saying, that decisions to change those principles, frameworks, and lexica must be subjective.

2. Friedman (2001:40) elaborates.

3. These occur variously in Quine (1964, Chap. 2; 2006; 2008a and b).

4. Franco (note 16) adds a “related worry” that he shares with Colin Marshall, who in a review of KCG writes: “Given these definitions [of ‘subjective’ and ‘objective’], one might think that it is trivially true that, setting aside properties, empirical terms and concepts have subjective and objective sources in Goldberg’s sense. On any familiar sense of ‘empirical,’ empirical terms and concepts depend essentially on subjects’ linguistic and conceptual capacities (subjective sources) and on objects or sensations deriving from objects (objective sources)” (Marshall 2015). And yet I address this point directly (KCG, pp. 18–20) by explicating the role that “essentially” plays in my definitions. Marshall also worries that my definitions are disjunctive. This is curious. Besides explaining why the—inclusive—disjunction is needed, I also show that, as Franco observes (note 4), in many cases the inclusive disjunction becomes a conjunction.

5. Davidson himself (2001b, essay 13), otherwise “Quine’s faithful student” (2001b:29), thinks that Quine eliminates two dogmas of empiricism—the analytic/synthetic distinction and radical reductionism—only to leave the third dogma, scheme/content dualism. Quine (1981) agrees. As I explained above, scheme/content dualism is one formulation of Dualism. Davidson groups Kant, Kuhn, Carnap, and Quine together as his dualists just as I group them together as my Dualists. Franco (note 15) himself notes the historical connection between Quine and Kant via C.I. Lewis, whom Davidson (2004:237) also groups with them. So for this reason too Quine should be understood as a Kantian.

6. See Feest and Sturm (2011b and especially 2011a) for an overview.

7. Nevertheless, see Franco’s fascinating suggestion from Ian Hacking (2015:25) on how genealogy and geography relate.


Allison, H. (2004), Kant’s Transcendental Idealism: An Interpretation and Defense, revised and enlarged (New Haven, CT: Yale University Press).

Carnap, R. (1988), ‘Empiricism, Semantics, and Ontology’, in Meaning and Necessity: A Study in Semantics and Modal Logic, 2nd ed. (Chicago: University of Chicago Press), suppl. A (pp. 205–29).

——— (2003), ‘The Logical Structure of the World’, in The Logical Structure of the World and Pseudoproblems in Philosophy, trans. R. George (Chicago: Open Court), pp. 3–303.

Coffa, A. (2008), The Semantic Tradition from Kant to Carnap: to the Vienna Station (Cambridge: Cambridge University Press).

Davidson, D. (2001a), Essays on Actions and Events (Oxford: Oxford University Press).

——— (2001b), Inquiries Into Truth and Interpretation (Oxford: Oxford University Press).

——— (2004), Problems of rationality (Oxford: Oxford University Press).

Edgar, S. (2015), ‘Intersubjectivity and Physical Laws in Post-Kantian Theory of Knowledge Natorp and Cassirer’, in M. Friedman et al. (eds), The Philosophy of Ernst Cassirer: A Novel Assessment (Berlin and New York: de Gruyter), pp. 141–62.

Feest, U. and T. Sturm (2011a), ‘What (Good) is Historical Epistemology? Editors’ introduction’. Erkenntnis 75: 285–302.

———, (eds) (2011b), ‘What (Good) is Historical Epistemology?’, special issue, Erkenntnis 75(3).

Faulkner, W. (1994) Requiem for a Nun. William Faulkner: Novels 1942–1954, ed. J. Blotner and N. Polk (New York: Library of America).

Foucault, M. (2007), Discipline and punish (New York: Vintage Books).

Friedman, M. (2001), Dynamics of Reason (Stanford: CSLI).

Gibson, R. (ed.) (2008), Quintessence: Basic Readings From the Philosophy of W.V. Quine (Cambridge, MA: Harvard University Press).

Hacking, I. (2015), ‘Let’s not Talk About Objectivity’, in F. Padovani et al. (eds), Objectivity in Science: New Perspectives From Science and Technology Studies (New York: Springer), pp. 19–34.

Kuhn, T. (2012), The Structure of Scientific Revolutions (Chicago: University of Chicago Press).

——— (2002), The Road Since Structure, ed. J. Conant and J. Haugeland (Chicago: University of Chicago Press).

Marshall, C. (2015), Review of Kantian Conceptual GeographyNotre Dame Philosophical Reviews:

Quine, W.V.O. (1964), Word and Object (Cambridge, MA: MIT Press).

——— (1981), ‘On the Very Idea of a Third Dogma’, in Theories and things (Cambridge, MA: Harvard University Press), pp. 38–42.

——— (1995), From stimulus to Science (Cambridge, MA: Harvard University Press).

——— (2004), Pursuit of Truth, revised ed. (Cambridge, MA: Harvard University Press).

——— (2006), ‘Two Dogmas of Empiricism’, in From a Logical Point of View: Nine Logico-Philosophical Essays, 2nd rev. edition (Cambridge, MA: Harvard University Press), pp. 20–46.

——— (2008a), ‘Truth by Convention, in Gibson (2008), pp. 3–30.

——— (2008b), ‘Carnap and Logical Truth’, in Gibson (2008), pp. 64–90.

Pettit, P. (2002) Rules, Reasons, and Norms: Selected Essays (New York: Oxford University Press).

Ryle, G. (2000), The Concept of Mind (Chicago: University of Chicago Press).

Sellars, W. (1968), Science and Metaphysics: Variations on Kantian Themes (New York: The Humanities Press).

——— (1991) Science, Perception and Reality (Atascadero: Ridgeview).

Schlick, M. (1979), Critical or Empiricist Interpretation of Modern Physics?’, in Philosophical Papers, vol. 1, 1909–1922, ed. H. Mulder and B. Van DeVelde-Schlick, trans. P. Heath (Dordrecht and Boston: D. Reidel), pp. 322–34.

© Nathaniel Goldberg, 2016.

Nathaniel Goldberg is Associate Professor of Philosophy at Washington and Lee University, Lexington, Virginia. He researches at the intersection of epistemology, metaphysics and philosophy of language, and has an abiding interest in Kant.