American Realism in the Early 20th Century

Invited Speakers: Direct Realism - Historical and Systematic Perspectives

Ralf Busse, Johannes Gutenberg University Mainz

Direct realism in perception: Kant vs Sellars the Younger

Wilfrid Sellars reports that his “new realist” father had him study Kant's Critique of Pure Reason as a young person. Both this reading and Roy Wood's anti-pragmatist realism left deep traces in the younger's philosophical work. On a closer look, however, Kant proves himself to be the more decided realist in important respects, and in his theory of perception in particular. With great justice, Sellars insists on the depth and importance of Kant's dualism of sensibility and understanding and highlights their team play on the "interesting meeting ground" of productive imagination. First, however, he incorrectly downplays the "causal theme" in Kant's account of sensibility, ending up with a more Humean than Kantian view of sense impressions. But Kantian impressions of outer sense are essentially that, im-pressions originating from an outer causal source, rather than qualia-loaded internal mental states that are intrinsically neutral with respect to their kind of causal origin. Secondly, Sellars wrongly reserves the achievement of intentionality to the conceptual level of the understanding, which leads him to an interpretation of Kant's outer intuitions as non-sensory, conceptual representations. For Kant, by contrast, outer intuitions are purely sensory representations that have their own kind of non-conceptual, purely causal intentionality in relation to the existing objects that are their causal sources. Thirdly, Sellars interprets productive imagination as a conceptually guided construction of inner quasi-sensory "image models", which stand proxy for outer objects in the "manifest image" of the world. For Kant, however, the non-causal, conscious part of perceptual intentionality does not so much consist in the conceptual content of "perceptual takings" per se, but in a mental practice of expecting coherent sequences of sensory inputs. It is by expecting a rule-governed coherence or "synthetic unity among given representations" that we consciously direct given intuitions to the common object that is, as a matter of fact, their causal source. In sum, Kantian outer perception is in both its receptive and its spontaneous aspect firmly directed at actual objects in the external world. Ch. S. Peirce's (the pragmatist!) interpretation is not far from truth that according to Kant, we directly perceive things in themselves (though not as they are in themselves).

Mazviita Chirimuuta, University of Edinburgh

Critical Realism and the Humanist’s Religion

This paper will explore the wider motivations of Roy Wood Sellars’ critical realism. In a chapter entitled “The Humanist’s Religion”, Sellars (1918:217) writes that a solution to the mind body problem is “imminent” and that this will spell the end of the “old supernaturalism” that shaped all previous religions. He refers to his earlier publications Critical Realism (1916) and Essentials of Philosophy (1917) for details on his non-dualist, non-reductive physicalist theory of mind. This indicates that the agenda behind Sellars’ philosophy of mind is not narrowly theoretical, but is deeply implicated in Sellars’ understanding of religion and concerns for the development of American society post WW1. This paper will examine how critical realism relates ordinary perceptions to scientific knowledge, and how this connects with Sellars’ “marriage of naturalism with humanism” – his aspirations, following Huxley, for the scientific methods to be applied to all “problems of life” (p.219). We will also see how this scientism relates to Sellars’ advocacy of socialism, as “both a science and an ideal” in The Next Step in Democracy (1916:24).

Christian Damböck, University of Vienna

Carnap, Russell, Perry: Chicago 1938 and Harvard 1940

Rudolf Carnap (here pretty much on a line with his Vienna Circle colleague Otto Neurath) always rejected strong forms of scientific realism. On the other hand, his semantic understanding of logical empiricism as it was developed against the background of discussions in the Vienna Circle is certainly a variety of a “realist” worldview being centered around the idea of objective knowledge that becomes possible by means of physicalism and a physicalist thing language. It is not only for us as historians often hard to carry out in which way this philosophical stance converges and diverges with scientific realism. Also, there were quite intense exchanges between Carnap and Russell, and also between Carnap and Perry, about the relationship of their philosophical standpoints. First, in 1938/39 Russell visited the University of Chicago and delivered a seminar that was attended not only by graduate students (such as Irving Copi and Abraham Kaplan) but also by Rudolf Carnap and Charles Morris. The basis of this seminar was Russell’s manuscript for his Inquiry into Meaning and Truth and the entire discussion was framed by an exchange between Russell and Carnap on the realism topic. In 1940/41 Carnap and Russell met again at Harvard on occasion of a visiting professorship of Carnap’s. Carnap organized two lecture series, on logic, and on philosophy of science. Among the prominent speakers and regular participants was also Ralph Barton Perry with whom Carnap also had several private meetings and intense philosophical discussions. This paper will reexamine these two episodes being important for the history of analytic philosophy. Beside of the existing secondary literature (e.g. a paper by Gary M. Slezak and Donald W. Jackanicz on the Chicago seminar and Greg Frost Arnold’s book on the Harvard episode) my main source will be the newly transcribed diaries of Rudolf Carnap who shed new light on these episodes. Moreover, I will also examine the relevant correspondence on this topic, between Carnap and correspondence partners such as Russell, Neurath, Feigl, Hempel, and Reichenbach.

John Dorsch, University of Edinburgh

Direct Perception, Critical Realism and Ecological Psychology

As far as it concerns perception, Roy Wood Sellars' case for direct realism shares a strong family resemblance with recent appeals in the philosophy of cognitive science to non-representational forms of cognition, specifically those that evoke Gibsonian affordances. One example is the affordance for climbing, argued to be directly perceived in virtue of an agent's active engagement with an appropriate distal environment (e.g. rocky slopes). Here, affordances play the cognitive role of perceptual representations, motivating and guiding behavior, but affordances are argued to be external properties of the mind-independent world, which the subject perceives directly in virtue of her capacities for action. Clearly then, such claims are similar in tone to the non-representational project advocated by Sellars' work on direct realism. Notice, however, that, so construed, Gibsonian affordances, while succeeding in explaining crucial aspects of cognition without recourse to internal mental representations, do so at the cost of an expansion in ontological commitment: affordances are external features of the mind-independent world, similar to other features detected through perception, such as shapes and colours. That said, little is known about the exact notion of directness operative in Sellars' case for direct realism, whether it is distinct from or conceptually similar to the notion at the heart of Gibsonian ecological psychology. Does Sellars’ notion of directness offer philosophers of cognitive science an alternative approach to making sense of non-representational perception without thereby requiring they expand their ontology to include e.g. 'apt for climbing'? By a close examination of Sellars’ philosophy of perception, I shall provide a comparative analysis of his notion of directness vis-à-vis Gibson's notion.

Roberta Locatelli, University of Tübingen

Naïve Realism and the Relationality of Phenomenal Character

Although naïve realism is becoming increasingly popular, what exactly naïve realists need to commit to remains often unclear.  Two claims are often used to specify the view:

Constitutive Claim (CC): The phenomenal character of perceptual experience is (partly) constituted by the mind-independent objects in one’s surrounding and their properties.

Relational Claim (RC): Perception is a relation to mind-independent objects in the environment and their properties.

Both claims are open to different interpretations and so far naïve realists haven’t made enough to clarify how they should be understood respectively. Moreover, it is not clear how they relate to one another. While some seem to think that naïve realism is committed to both CC and RC, some have tried to disjoin them: Beck (2018) argued that only RC form the core commitment of naïve realism, which would be better off denying CC, while Steenhagen (2019), on the contrary, that only CC does.
This paper argues that CC and RC are both integral parts of naïve realism. In particular, it is not possible to offer an intelligible and plausible understanding of CC other than as a corollary of RC. After surveying various possible interpretations of CC and RC and showing that they are all either trivial or widely implausible, I advance an understanding of both claims in which they are inextricably intertwined.
What’s central to the naïve realist’s understanding of phenomenal character is that they refuse any account of the phenomenal character in term of special properties (be them properties of the experience itself like qualia, mental entities like sense data or representational contents). What explains the phenomenal character is simply the world-involving fact that a subject has a point of view on the world. The phenomenal character of an experience captures what it is like to have an experience. And seeing a red apple is just like being presented with red apple from a particular perspective.
The phenomenal character is not an entity or a special property of the perception, but just a particular way of looking at the relation of awareness that perception is: how it strikes the subject undergoing it. Therefore, RC is not only a claim about the nature or structure of perception, independent of claims about the phenomenal character. So understood, RC claims that the phenomenal character of perception is identical to the obtaining of the relation of having a point of view on an object.
We should understand RC as a claim about the phenomenal character of perception, rather than as a claim about the nature or structure of perception, as it is most often understood. So understood, RC claims that the phenomenal character of perception is identical to the obtaining of a relation to an object.
In turn, this allows us to see CC as a corollary of RC: it is because the phenomenal character of experience is itself relational: it is the obtaining of a relation to an object, and relations can’t obtain in the absence of their relata, that the object is a constituent of the phenomenal character.
This interpretation of what is the core commitment of naïve realism has significant implications not only because it puts us in a position to evaluate naïve realism (we can’t assess a view without understanding what it commits us to), and understand to what extent recently proposed versions of naïve realism that try to amend the view in the light of recent criticism, succeed in preserving its original motivation. Most importantly, the paper shows that the disagreement between naïve realists and their opponents lies in the different ways they understand the phenomenal character: while opponents have explained it in terms of special entities of one sort of another, naïve realists explain it entirely in terms of the relation of awareness that obtains when we perceive.

Sofia Miguens, University of Porto

Seeing Things As They Are and Seeing-As
(Searle, McDowell and Travis)

The claim that all perception is representational, by which it is meant that all perception is of something under an aspect, is a central part of John Searle’s 2015 formulation of direct realism. In this talk I will resort to both John McDowell and Charles Travis in order to criticize different points of Searle’s claim. My ultimate interest is the contrast between McDowell and Travis themselves where it concerns phenomena of seeing-as.

Matthias Neuber, Johannes Gutenberg University Mainz

Russell and American Realism

American philosophical realism emerged and developed in two forms: "new" and "critical" realism. While the new realists (R. B. Perry, W. P. Montague, E. B. Holt, and others) sought to 'emancipate' ontology from epistemology and at the same time defended a direct theory of perception, the critical realists (R. W. Sellars, G. Santayana, A. O. Lovejoy, and others) explicitly promoted a representationalist account of perception and thus argued in terms of an "epistemological dualism" (Lovejoy). Bertrand Russell's early philosophical writings figured prominently in both of these American realist camps. However, while the new realists quite enthusiastically embraced the Russellian analytic style of reasoning (and Russell himself appreciated the American new realists as anti-idealist allies), the critical realists were significantly more reserved toward Russell's point of view. In the talk, I shall attempt to identify the reasons for this difference in reception. In particular, I will try to show that the critical realists' reservations pertained primarily to a more traditionalist – 'continental' – account of philosophy, whereas the neo-realists proved instrumental in setting the stage for later full-blown analytic philosophy in the United States. While the latter praised Russell as the pioneer of a new philosophical age ("logic as the essence of philosophy"), the former demarcated themselves from the analytic movement which displeased them as "a form of neo-colonialism" (Sellars). By spelling out the details of this inner-realistic tension a clearer picture of early twentieth-century American philosophy will, it is hoped, begin to emerge.

James O'Shea, University College Dublin

Critical Direct Realism? Roy Wood Sellars and Wilfrid Sellars

Against the ‘new realism’ or neo-realism of the early twentieth century, the critical realist philosopher Roy Wood Sellars (1880–1973) sought to retain the cognitive directness of perceptual knowledge in relation to physical objects while rejecting the neo-realists’ conception of a direct ‘acquaintance’ with physical objects – or as he put it, while rejecting the idea “that the object of knowledge is the character-complex of which we are aware” (1920: 189). His goal was thus to defend a middle way between Lockean ‘indirect realism’ as traditionally interpreted and the ‘direct acquaintance’ positions shared by naive realists, neo-realists, and idealists alike. This ‘critical direct realist’ theory of perceptual knowledge, as we may call it, was inherited and modified in certain ways by his son Wilfrid Sellars (1912–1989). In this paper I explore this historical comparison in order to shed light on what sort of position emerges from this line of development, and to assess its plausibility.

Jani Sinokki, University of Oulu

Intentionality and Objective Identity in Descartes

Descartes attempts to explain the intentionality of ideas by stating that ideas have or contain objective reality. This objective reality, in turn, is said to be identical to the reality of the object of the idea, so that ‘the idea of the sun is the sun itself existing in the intellect’. This view has been interpreted as amounting to a content approach to intentionality, implying that Descartes is, contrary to the received view, a direct realist. I will challenge the first part of this assumption and argue with the traditional readings that Descartes’ view does not amount to a content approach, but instead to an object approach to intentionality – the objective reality of an idea is the proper object of awareness itself. However, contrary to what is often assumed, I show that understood in the light of his commitment to sortalism, Descartes’ view seems nevertheless more consistent with direct realism than indirect realism. After all, Descartes is claiming that the objects in our ideas are identical to what exists outside us. I argue that this should be understood in the same sense as a stretch of DNA that is in me can be numerically identical (in the sortalist sense; qua a stretch of DNA) to the stretch of DNA in my parent despite those particular stretches being in another sense distinct stretches. Interestingly, given Descartes’ commitment to a causal theory of perception, the theory I am attributing to Descartes seems to fall in between direct and indirect realism while bearing earmarks of both, but arguably should still be classified as direct realist.

Charles Travis, King's College London

Knowing The Ins And Outs

“But it must be like this” is not a proposition of philosophy.

(Wittgenstein, Philosophical Investigations, §599)

There is a very large part of John McDowell’s philosophical work that I can but first admire, then apply. To borrow a phrase from Wittgenstein, I think he has shown us just how, most productively and insightfully, to rotate our enquiries around the axes of our real needs. For all of which, there is one area of his work which he and I have been debating for, at this writing, 18 years. Perhaps I misunderstand what he is saying. But in any case, after years of trying, I can neither agree nor fit it together with that large part of the work just mentioned. Now as, in his words, ‘time is shrinking’, (exactly) equally for both of us, I want to make one more try at saying what I think is wrong with it.
I am afraid this will not be short. It does divide, though, into two main issues. The first concerns the notion proof. As I read McDowell, proof is essentially a matter of truth-transmission, beginning from, as Frege put it, ‘already acknowledged truths’, or better put for McDowell’s purpose, (prima facie) to be acknowledged truth. This idea occurs, or at least is put to serious work, initially in his Locke lectures, now nearly thirty years old. But I cannot find the place where it is renounced, or at least ceases to be put to serious work. So it will be central here.
The second idea up for critical attention is, in brief, that a philosopher, or at least one who pretends to treat a capacity of ours, e.g., our capacity for knowledge, perhaps restricted to our (spatiotemporal) environment, is due us an account, of some sort of other, of how such a capacity could be enabled, an account in which, most likely, there is going to be reference to specific sorts of enabling occurrences, or processes. Such an account need not be, or at least be advertised as, work for empirical psychology. It may, in fact, advertise itself as merely an investigation into how things would have to be for anyone who enjoyed the sort of unbounded capacity for knowledge creatures like us in fact enjoy, thus as applicable to martians or Douglas Adams’ dolphins, should either of such there be—the sort of being sketched by Descartes in 1637. (Thus, in Wittgenstein’s slogan above, an enabling ‘must’, part of that ‘proof of P’ attributed in some quarters to Sidney Morgenbesser: ‘If not P, what? Q maybe?’.)
Here, I think, we have a case of the most important divergence between Kant and Frege. I use the name ‘Kant’ advisedly. I mean Kant as he has in fact influenced, inter alia, the ‘anglosphere’ (term thanks to Emmanuel Macron). Kant’s interest in enabling, as I understand it, was an ingenious and original response to the challenge posed by the British empiricists, as they cast their eyes over the carnage and destruction in 17th century England wrought by the rise of the ‘flashed upon’ route to knowledge (and conspicuously on the rise to day in America, as in ‘There are paedophiles under that pizza parlour’, ‘How do you know’, ‘It flashed on me’. Enough, almost, to make  one sympathise with the empiricists. But it is enough here for a philosopher to recall how beneficial Frege showed us it may be to ask first, what phenomena form the object of the supposed capacity—what there is, in fact, for such a capacity to do, or be, or recognise—and only later how it might be enabled, if, having come so far, the questions remaining seem in any way philosophical at all.
The purportedly two questions I have distinguished here may be hard, on such short acquaintance, to recognise as two rather than one. Questions about proof might be seen as eo ipso belonging to the business of enabling, rather than as ones as to  the nature of what there is to be enabled.  Such is one reason why I cannot claim to be constructing here a counter-example to the principle that nothing is shorter than it is. But by the end, I hope that, for the by-then-long-suffering the distinction will be clear enough.

Ádám Tamás Tuboly, MTA Lendület Values and Science Research Group


Krisztián Pete, Department of Philosophy, University of Pécs

A Sellarsian Solution to the Explication Problem: Carnap in the Space of Reasons

Logical positivism is known for many things, but one of the most interesting points of reference is the relation between its anti-metaphysical stance and sort-of direct realist sense-data theory of knowledge. The two seem to contradict each other, and much ink has been spilled on this issue from the 1950s to recent times. Among the critical voices, one of the loudest was that of Wilfrid Sellars. In our paper, we would like to take a fresh look on the Carnap-Sellars debate since the majority of the reconstructions of their exchange embrace and build on the disagreement regarding realism, directness, inferences, and their relation to ontology and semantics. Our intention is to converge the two philosophers. We aim to take a look at Carnap’s explication pointing out some difficulties regarding the alleged conservativeness of the process of explicating ambiguous ordinary terms and claim that by borrowing some ideas or approaches from Sellars, we can fill in the gaps in Carnap’s story. André Carus has recently examined Sellars’ critique of Carnap’s position, but in our view, Carus misrepresented Sellars’ views for the sake of distancing him from Carnap. Thus, we will argue that a refined understanding of Sellarsian ideas regarding the directness of perception, the status and nature of the Manifest Image could provide important and fruitful solutions to some of the basic problems of explication, especially about the relation of explicandum and explicatum.