Reappraising the Notion of Medium through an Interpretation of the Philosophy of Wilfrid Sellars

SILVIA MOLLICCHI

Ada Lovelace Institute, UK

 

Abstract

This article argues that Wilfrid Sellars’ work has implications for the notion of medium (and media) and how we discuss it in cultural and media theory. The research introduces Sellarsian materialism, which is both non-dualist and non-reductionist with respect to conception, in the context of Media Theory and traces within it an otherwise implicit articulation of the notion of medium that, is argued, becomes necessary when accounting for the working of rules of validity in a manner that does not naturalise the normative. The article, which begins by putting Sellars in dialogue with authors influenced by Kittler’s work, mobilizes Sellars’ writings on language acquisition and epistemology to show how a notion of medium emerges in Sellarsian philosophy of language, as a consequence of a non-dualist account of validity in linguistic behaviour. This notion explains the relation between language and conception: language can be described as the medium of conception, as opposed to being its means of expression or identified with conception.

Keywords

Medium, epistemology, ontology, media theory, metaphysics, Wilfrid Sellars

 

Introduction

In the philosophy of Wilfrid Sellars, language, thought and conceptualization are positioned in such a way that Johanna Seibt (1990: 119) has described language in Sellars as the medium of conceptualization, a description that this article takes to mean that language is that in which thoughts come to appear to us as concepts. As I will show, this description revolves around a specific notion of medium, defined as not epistemologically separable from what it mediates nor ontologically equal to what it makes appear.

This broad notion of medium implicitly emerges in Sellarsian epistemology as necessary to account for the normativity—here understood in contraposition to the natural and presiding over questions concerning ‘what ought to be’ rather than ‘what is’—which uniquely characterises conception, in a manner that does not commit us to abstract entities—historically charged with grounding conceptions (Sellars adheres to a strict nominalism). In other words, when Sellars explains how rules of validity, the implicit commitment to which constitutes conceptions, hold sway over our purview of the world—that is to say how logically a priori norms, informing the transcendental level that enables our experience of the empirical level, are instituted formally rather than temporally, such that the two levels are always already differentiated—a notion of medium becomes indispensable.

In the context of theorisations of media, notions of the transcendental, epistemology and epistemic conditions are referred to, studied and confronted by a variety of authors, among whom we could mention those who have written about Cultural Techniques (Siegert, Krämer, Bredekamp, Macho) and Media Archaeology (here I refer to Ernst), and in the work of Kittler. While these thinkers would share Sellars’ commitment to the fact that only concrete particulars exist, and therefore any explanation of the working of rules of validity (and of conception) should avoid abstractions, they do not foreground said explanation in a manner that accounts for the power of norms—something which Sellars can instead supply. For this reason, the encounter between Sellarsian philosophy and the media theorists referenced above seems especially productive. Not only can Sellars be read as sharing some of their concerns, but the solution offered to those concerns is undergirded by a notion of medium, which contributes to an account of the formal character of the transcendental from within a theorization of the medium. This theorization advances reflections on present modes of knowledge, on how non-linguistic media may condition experience—with or without being first mediated by language and conception—and on what that might mean for what parts of reality come to appear to a subject and for how subjects are shaped.

The claims therefore are that: theories of the medium ought to account for the formal character of the transcendental and of rules of validity and that a Sellarsian notion of medium, precisely because it is introduced while accounting for validity, not only warrants said account, but also places the medium at the core of questions related to the transcendental. Furthermore, adopting this notion, which defines the medium in rapport with questions of knowledge and ontology, rather than as the general term for different classes of media (technical media, mass media, cultural techniques, etc.) could enable us to refine questions on the role of media in conditioning our interactions with the world in relation to conception or separate from it. The first two points constitute the core of the article, while the latter one remains implicit and will require further work.

The present article introduces Sellarsian philosophy to the readers of media theory. For context, the main headlines of Sellars’ complex philosophical project—which has been read from a variety of perspectives according to a pragmatist epistemological as well as a metaphysical realist point of view (see for instance, Brassier, 2007; 2011; 2013; Delaney et all, 1977; O’Shea, 2007; Rosenberg, 1998; 2007; Sachs, 2014; Seibt, 1990 etc.)—that are relevant to the present article are that, for Sellars:

[t]he 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. Under ‘things in the broadest possible sense’ I include such radically different items as not only ‘cabbages and kings’, but numbers and duties, possibilities and finger snaps, aesthetic experience and death (1962, para.1).

In order to understand this hanging together, Sellars develops a double methodology. We look at the world, and our place as limited knowers in it, according to two modalities: one is correlational and constitutes what Sellars calls the Manifest Image, and one is postulational and constitutes the Scientific Image. In the context of the second image, we attempt to formulate questions of the order of what is, while in the context of the first image, we attempt to grasp our commitments to what we supposedly sustain ought to be. Notably, we approach the second image always through the first one (our epistemic sphere), but this does not mean that we cannot genuinely aim at addressing its questions. In other words, Sellars articulates the normative (what ought to be) and the natural (what is) spheres in a manner that tries to square rationalist, empiricist and pragmatist traditions, maintaining that the normative cannot be deduced from the natural, but the natural obviously exists and we can try to improve our knowledge of it. In concise fashion, we may say that Sellars abides to a combination of normative anti-naturalism and ontological naturalism. All this is obviously far afield from any familiar discussion of the notion of medium in media theory. However, from the perspective of Sellarsian philosophy, the non-ontological character of the normative raises questions as to how norms, which are not ontologically grounded, come to appear. Thus, although the notion of medium is not a characteristic one of the Sellarsian vocabulary, it may become necessary, as I will argue in the rest of the article. From the perspective of media theory, it is advantageous to trace the contours of a notion of medium that is explicitly grounded on the rejection of an ontology of norms (and their possible naturalism) and emerges at the point when we explain how norms come nonetheless to have effects. Doing so may contribute to avoid both the risk of ontologising concepts and naturalising what subjects supposedly do or ought to be, while we theorise the role that media of various kinds may have in producing them.

In the article, I initially frame the introduction to Sellars as a conversation with authors influenced and somewhat continuing Kittler’s work and their stance vis-a-vis epistemology. The article then delves into Sellars’ work, starting from his justification for the necessity to account for validity—and doing so in the context of language—and continuing through the explanation of how we commit to certain world-views as opposed to others. The notion of medium is introduced to categorise how language is positioned vis-a-vis conception and thought, and further explained in terms of how it helps articulate ontology and epistemology in Sellars’ special brand of materialism. The article reflects on this specific appeal to the notion of medium and draws out the generic definition of it that we can trace in Sellars’ work, following Seibt’s interpretation.

In the paper, a specific medium, language, is named as the medium of conception. However, the argument does not focus on the fact that language is here defined as a medium. The paper revolves around the notion of medium, which comes about when the power of norms and our capacity to know them is explained by establishing a specific relation of mediation between language, conception and thought, as opposed to a relation of expression, where the mediated would be voiced out through the ‘tool’ of the medium, or of identity between medium and mediated.

 

Setting the distinction between transcendental and empirical

In 2013, an introduction to the concept of Cultural Techniques as one of the trends inheriting the legacy of what is generally called German Media Theory was published for the benefit of the English academic audience in the form of a collection of articles (Winthrop-Young, Iurascu and Parikka, eds., 2013). The publication featured various perspectives, including one committed to the method of Media Archaeology (Ernst, 2016; 2013), which is in tension with Cultural Techniques, but shares a similar Kittlerean influence. We can take these three positions—Cultural Techniques, Media Archaeology and earlier German Media Theory—as possible interlocutors within the camp of the study of media for the reflection on the notion of medium I here advance.

Citing from the collection, Siegert’s ‘Cultural Techniques: Or the End of the Intellectual Postwar Era in German Media Theory’ (2013) traces the relation between the German Media Theory of Kittler, Theweleit, Bolz, Tholen and Siegert himself and the more recent trend of Cultural Techniques, and constitutes a fitting starting point to explain how Sellars’ work may intersect the study of the medium. If not touched upon explicitly, the language of Siegert’s article implies the fact that the reflection on technical media and cultural techniques in Germany have inscribed the study of specific, historical media in a discourse that circles back to underlying motives of the transcendental, the distinction between transcendental and empirical levels and, with them, the articulation of epistemology—understood as the domain of knowledge, through which we access reality in a manner that is formally demarcated by the transcendental, which sets the conditions of possibility of experience—and ontology—understood as the discourse on what is.

Siegert’s account begins by describing the conflictual context in which a relatively small group of German scholars, profoundly influenced by (as well as responding to) French post-structuralism, turned their attention to marginal and oft-neglected technical media (rather than mass media of communication). In the 1980s, these scholars began to work with an anti-philosophical, and yet profoundly theoretical, implicit notion of ‘medium’. The conflict in question, which characterised postwar German intellectual life, Siegert writes, resulted in a show-down between this group and various offshoots of the Frankfurt School, including cultural studies. Media and culture were competing over the ‘vacant throne of the transcendental’ (2013: 49). Notably, the initial revolt against mainstream humanities and its intention to kick the Spirit (Kittler, 1999) out of them was animated by a firm anti-humanism that both Cultural Techniques and Media Archaeology have continued with to a certain extent or tempered in the form of post-human discourse. That media had risen to the throne of the transcendental, according to Kittler, meant that the determining role of material media on perception, on the episteme we live through and therefore on all human activities (1990; 1997; 1999) had to become object of studies for the humanities also in order to overcome their then common interest, the human, and its philosophical and cultural formations (1990: 286).

We could read Media Theory (and specifically Kittler), the study of Cultural Techniques and Media Archaeology in terms of their respective relations to epistemology.

Kittler’s anti-humanist position is, effectively, an anti-epistemological one. If we were to apply Kantian terms to Kittler—who rarely makes explicit reference to Kantian transcendental philosophy—we may say that material media takes over the throne of the transcendental, but they do so by eliding the distinction between empirical and transcendental levels. Kittler marshals a reductionist programme with regards to conception and, hence, normativity, which characterise the transcendental level and its formal distinction from the empirical. In the absence of this distinction, epistemology collapses onto ontology. This position has been described as naive and implicitly re-inscribing the figure of man back into the picture. Kittler’s notion of media determining our perceptive faculties has been indicted for deriving the transcendental level from the empirical one, a crass mistake, since instead of explaining how the transcendental holds sway on a subject’s purview of the world, it ignores its normative force and keeps it (and Man) alive, but treacherously implicit (Sutherland and Patsoura, 2017: 61-2). There is of course another explanation for Kittler’s position, which is that he was drawing attention to the inevitably anthropological core of Kantian and post-Kantian transcendental philosophy, in which a specific notion of human being is always already determining of a specific notion of the transcendental. This is the empirico-transcendental doublet that Foucault identifies (Balibar, 2015: 63; Foucault, 2002; 2005: 448-9; and 2008), but somewhat recognises as inevitable (Sutherland and Patsoura, 2017: 58; Foucault, 2002). For Kittler, in contrast, technical media constituted a transcendental that was supposedly not drawn out of human anthropology and was therefore uncontrollable. However, the anti-humanist stance that characterises his work, by simply denying the formal character of the transcendental and omitting an explanation as to how any kind of transcendental, including a media-based one, would instal its rules of validity and have one implicitly commit to specific norms and activities, quietly leaves the human ways of conceiving at its core and falls into contradiction on its own terms.

The research on Cultural Techniques operates according to different terms. The context—no longer defined by the opposition to a techno-sceptic, humanist tradition (see Siegert, 2013: 55)—and the objects of study—from technical media to human practices of cultural techniques, which structure subject-object relations (Siegert, 2013: 56-58)—have changed. The rift between material media and culture has been sutured and the relations between cultural techniques, human culture and conceptual formations are explicitly at stake. However, among various of its proponents, Cultural Techniques are posited as preceding nature/culture distinctions, or, in a language amenable to the one used for Kittler in the previous paragraphs, empirical and transcendental levels. The symbolic (which we may relate to the conceptual) character of Cultural Techniques is acknowledged (Macho, 2013: 31, 44), but it remains unclear what characterises it and how it relates to their concrete character. In Macho’s writing, Cultural Techniques are said to be prior to the concepts they are related to (2003; 2013: 44) as well as to their actual media (2008: 44). This turn of phrase betrays a temporal understanding (45) of conception in relation to the empirical and again obliterates the normative level, at which concepts operate, that is neither temporally prior nor posterior to the empirical, but simply organises the access to it at a different, non-empirical level. Krämer and Bredekamp seem to commit to a similar obliteration (2013: 21). In their programmatic contribution to the 2013 collection, while differentiating symbolic and technical dimensions of culture and stressing the import of the latter, they avoid a treatment, in media-terms, of the normative dimension of culture and its inevitable being at work in any epistemic inquiry.

While occupying a different camp, Ernst’s account of Media Archaeology appears affected by the same type of obliteration. To make an example, signal-processing machines (2013: 140), which are culturally generated, detach from culture and the symbolic sphere and stop a semantics-based mode of processing, according to Ernst. They develop separate means of interaction with other machines and code-based systems. However, Ernst here seems to avoid the fact that whatever we believe we see in signal-processing machines, we still observe it from within the transcendental condition we inhabit. A writer of Media Archaeology may emphasise the partial (or even complete) autonomy of a machine and how it re-constitutes time, however they are still not writing from the perspective of the machine, but from within semantics and according to the default sense of time that is constitutive of an all too human transcendental.

A Kantian critique, similar to the one moved against Kittler, could be moved against the referenced positions in the contexts of Cultural Techniques and Media Archaeology, as their implications are that something that is transcendentally determined (how we grasp media of all sorts) is somehow generated by an empirical entity and grasped outside presiding systems of norms that regulate our perception of reality. Again, in these cases as in Kittler’s, we may be applying a Kantian critique to non-Kantian positions, which only (and commendably) aim at overcoming the bind between a notion of human and a specific kind of transcendental. However, even if those in question were non-Kantian stances, they may still be in danger of contradicting themselves and omitting an explanation of how validity works, that is how experience is conditioned in a manner that is formally and not temporally a priori. This account could be given in terms of the a priori character of a medium, but one which operates exclusively in terms of the medium in question, without any support, not even linguistic, that we may use to discuss said medium, and barring any possibility of even imagining what was prior to it. Alternatively, the account of validity will include, in some way, the one of a priori norms and of how conception works.

To summarise, first of all, the commendable rejection of the ontology of philosophical concepts, which is proper to both Cultural Techniques and Media Archaeology, is endangered if a fully non-ontological account of concepts is not provided. Second, it is necessary to account for one of the most distinct traits of whatever makes access to the world for the way it is perceived possible, the fact that when one acts in the world or uses a word to label an activity they are also, apparently, aware of whether that use is valid or not—here validity is understood as what we attribute to behaviours that supposedly comply with socially established norms.

This is the point at which Sellars’ philosophical work may constitute an unlikely ally for those studying media within a post-Kittlerean tradition. Sellarsian philosophy is preoccupied with the distinction between the transcendental and the empirical levels and the articulation of epistemological and ontological domains and, as he works on these distinctions, according to Seibt’s reading, a broad notion of medium emerges as necessary to set them. This notion is fleshed out in the process of accounting for a-priori norms and, hence, conception in a manner that is rigorously non-dualist and does not ontologise concepts.

Here an objection could be raised: the attribution of the status of medium to language, which is part of the argument, is a hotly debated topic in the context of theorizations of the medium (for opposers of the idea that language is a medium see, for instance, Ernst, 2016; and for supporters see Benjamin, 2004: 62-74; and Weber, 2008: 31-52; see also Guillory, 2016 for a general history of the debate). The proposal I advance to use Sellars to rethink the notion of medium, however, remains pertinent regardless one’s position vis-a-vis this debate because the path that, through Sellars’ writing, identifies the notion of medium does not begin by stating that language is a medium. The label of ‘medium’, which Seibt references as preceding any identification of the relation between language and conception, is used to classify that relation once its characteristics have been demonstrated. Sellars does not start with the problem of the medium, but with the issue that the authors of cultural techniques seem to by-pass, the one of accounting for rules of validity, which, in Sellars, constitute conception, that hold sway on our interactions with empirical reality, its involvement in our perception, including of human practices, objects and cultural techniques. It is telling that, in addressing this issue, a notion of medium emerges, one through which the mediation (more than the generation) of concepts (if concepts have no ontological dimension, then they will not be generated, but they can still be mediated) is accounted for and can be premised to the mediation of technical media and cultural techniques. Sellars and the study of media meet half-way in a notion of medium that is deduced in order to both dissolve any hint at a dualist nature of concepts and account for how technics and techniques may be mediating what, of reality, appears to a subject, including the conscious and non-conscious acts that differentiate subjects and objects. The encounter I here propose offers, on the one hand, an account of the formal character of the transcendental and of knowing from within a theory of the medium and, on the other hand, a notion of medium at the intersection of questions of ‘what is?’ and ‘how do we know?’, which may first be asked in connection and with reference to language and then to any medium.

Before moving onto the discussion of Sellarsian philosophy, it is important to clarify that Sellars differs from the authors of Cultural Techniques and Media Archaeology with regards to their position towards anti- and post-humanism. Sellars is external to the debate around anthropology and post-humanities. Is this discrepancy meaningful in the context of this discussion? Sellars does not raise issues concerning the empirico-transcendental doublet, the necessity of de-centering the human in the context of epistemology and the self-coherent figure of the transcendental subject in these very terms, but his work may be read in a manner that helps us think through all of these concerns. Pertinently, Sellars attempts a necessary account for the subject and its indirect (and hence potentially mistaken) access to empirical reality in a non-naive manner—a commitment that does not inherently contradict the objective of overcoming anthropocentrism. Once we have deduced whatever notion of medium is implied in Sellars’ method of articulating epistemology and ontology, we can attempt at formally reproducing the same structure in relation to other media while remaining epistemologically sound. What we are left with is a demonstration of the necessity of a support (a medium) upon which both conscious and non-conscious acts—with the latter being thoughts that appear in mediated conceptions or other practices—are impressed and which organises the alternate and impermanent distinctions between subject and object.

In the rest of the paper, this necessary medium will be deduced from a reading of Sellarsian philosophy and defined in relation to what is mediated in such a manner that the medium cannot be parsed from what is mediated, but is also different and distinct from whatever it makes appear. In the first instance, this notion enables us to account for the institution of validity. In the second instance, which would require a separate article, it furthers the reflection on how the rapports between conceptions (and language) and media feed into the co-constitutive processes of interacting with reality and shaping subjectivities.

 

No account of language can avoid one of validity

To understand how Sellars frames the broader problem of conception, which I have identified as implicitly shared with authors of Media Theory, Cultural Techniques and Media Archaeology, let us look at ‘Language, Rules and Behaviour’ (1949). Here Sellars lays the foundations of his critique of psychologist tendencies in then prevalent behaviourism (Ryle, 1949; Skinner, 1938). This critique will set the problem in terms of rules of validity, the working of which Sellars will then need to explain positioning language through a notion of medium, which we will consider in the next section. Once the notion of medium is introduced, we will look back at Sellars’ explanation and deduce what its characteristics are.

Sellars recognizes the methodological benefits of behaviourism (1978: para. 3-4), but rejects the idea that language could be explained only in empirical terms as a mere phenomenon of ‘tied behaviour’, whereby we respond to words as environmental stimuli that can be described through external observation. What logical and radical behaviourism cannot account for, according to Sellars, is the fact that we recognise when the use of a word is correct, and do so in contexts far removed from the original environment, in which we first apprehended a term qua sound and grapheme. A behaviourist approach to language cannot grapple with the question of validity and the working of the rules we assign it (i.e. conception). While justifying validity was not on the behaviourist agenda, according to Sellars, it is indispensable if we want to provide a plausible account of language use. Without showing how we define whether something is valid or not, whatever explanation we offer of language will not resemble how we patently use it. This is not to say that language necessarily works in the way that it seems to be working for us. It rather means that the fact that validity is crucial to our use of language cannot be left unexplained.

For Sellars, language is both a fantastically complex stream of noises that we learn to associate with environmental entities according to inputs we receive, and something that operates on the basis of rules that turn noises into symbols. The noise ‘blue’ “becomes a mediating link between what can suggestively be called a rule-regulated calculus, and a cluster of conditioned responses which binds us to our environment. Here we should note that the rules which inter-relate these mediating symbols qua linguistic symbols must mesh with the inter-relationships of these symbols qua tied symbols […]” (1949: para. 22, my emphasis).

‘Blue’ is a function within a structure, bound to coordinated rules, and it is in virtue of the fact that the grapheme or sound ‘blue’ is a mediating point between noise-to-respond-to-accordingly and function-to-calculate that we label it a word (1949: para. 28). The symbolic activities that sapient beings seem to display, Sellars sustains, are due to the extra-referential components of inference (those that don’t refer to the empirical reality we take in as a psychological fact). ‘Non-tied’ symbolic activities are, ultimately, extra-referential and, for this reason, can only be grasped in terms of rules of validity. The task at hand is explaining coherently this aspect of language without resorting to anything non-physical.

Clearly, talking about rules of validity and their being at work, and affirming that only concrete entities exist, are statements that may contradict each other. Moreover, the above tenet on the extra-referential character of inference risks activating a logical regress (itself typically Kantian). If we consider that the meaning of a term corresponds to its use in a language (and not an extra-linguistic entity), then learning to use a language would be equal to learning to obey the rules of that language. However, the rule for the use of a term would take the form of a sentence in the correspondent meta-language containing that term. Therefore to use a language one would first need to know its corresponding meta-language, something which is patently impossible (Sellars, 1991: 321; Seibt, 1990: 107 ff). To counter this circular argument, following the steps of Sellarsian epistemology, we need to prove that norms may be at work, without us knowing them, and that all we can do with norms is ‘coming to know’ them. Norms do not exist as such. Something else, entailing only concrete particulars, might exist and warrant the apparent regularity of our activities.

 

Leveraging on ought-to-be rules of state to stop the regress

To show the above, Sellars offers an account of language acquisition that lays the foundations for grappling with the relation between language and conception. To summarise this account, I refer to the texts where Sellars discusses language acquisition (see throughout 1953; 1969; 1973a; 1973b; 1991: 321-358) which revolves around the possibility of locating two epistemic positions, one that responds to ‘ought-to-do’ rules and one that responds to ‘ought-to-be’ rules.

Following Sellars, the person acquiring the language is responding to an ought-to-be rule or a rule of state. When seeing something red, one shall be in a ‘“red”-state’, as suggested through repetition by the language trainer, or in a ‘“door”-state’, or even in an ‘“and”-state’ according to a mode of parroting people who have already ‘acquired’ the language and speak it in line with the social norms of a community. The person transmitting the language, instead, is responding to an ought-to-do rule: they shall carry out an action that reinforces the compliance to the rule of state that the language trainee is responding to, such as choosing to use a certain term at a certain point, in coordinated fashion with other activities (Seibt, 1990: 111-8).

By ‘being in a certain state’ or responding to an ‘ought-to-be’ rule, I mean being in the state of ‘thinking-out-loud that X’. This ‘thinking-out-loud’ may cash into an utterance, but most of the time does not. Imitating patterns of behaviour, the language trainee does not simply become capable of uttering a certain term at a certain moment. They also begin to ‘absorb’ the linguistic structures that contribute to differentiating portions of reality, not in and of themselves, but in rapport with the habitual gestures we make when dealing with them, beginning to act in the world in a less-than-aware manner and according to effective social norms that we are absorbing. I take this to mean that beginning to recognize when one should be in a ‘door-state’ entails recognizing what doors generally look like and what one does in the presence of a door as well as when one may be using the word ‘door’ also in the case when no door is in sight (for instance when someone says, metaphorically, ‘revolving-door’). All these aspects are apprehended in synchronised fashion. Indeed, only when apprehended together with the notion of ‘opening’, ‘handle’, ‘next’, ‘room’, the language trainee begins to learn how to describe when one should be using the term ‘door’ according to the prevalent social norms. One’s capacity to eventually explain how a term is used and how it implicitly shapes one’s view of reality could be equalled to the capacity to draw a complex thread that translates the acts we carry out into linguistic terms.

The difference between the position of the language trainer and that of the language trainee is that the former can only parrot an existing pattern of use, while the latter, upon request, could offer an explanation as to why a certain word is used in a certain context (Sellars, 1974; 1973b: 489, 492). Obviously, the explanation may be incorrect, but the trainer can show, to some extent, how one term relates intra-linguistically to other terms and extra-linguistically to socially contextualised activities. Meanwhile, part of their role is to set up portions of reality appearing in front of the language trainee in such a way that the trainee’s use of a term in a certain circumstance will be reinforced or discouraged. Slowly, the trainee will be able to account for the use of a term. Again, this awareness will come about through the construction of webs of terms that relate to each other intra-linguistically. This is because language, and not autonomous mental entities, will be the vector coordinating the acquisition of normalised behaviour. The language trainee ‘thinks-out-loud’ and embeds in their uttering of words specific patterns, which, upon development of intra-linguistic relations among terms, they will also be able to criticise, explain to themselves and the rest of the world (Sellars, 1973a: para. 33; 1969: 96). This further skill, according to Sellars, is the mark of one’s capacity to access the meta-language. Importantly, while the use of linguistic utterances directs and contributes to structure our interactions with reality, the coordination of these utterances with physical sensation and consequential movements (material interventions in the world) plays a much greater role in shaping those interactions with any degree of continuity than mere linguistic norm compliance would warrant.

Notably, this account works not only for primary language acquisition, but for the use of language among fully-fledged members of a linguistic community. Even when we supposedly can explain the use of a term, our verbal behaviour is not parroting a rule, nor is it applying it with a precise degree of awareness, upon deliberation on complex rules of use. We mostly go about our lives using language without appealing to rules of use that we could plausibly spell out. Our behaviour is neither rule-following nor contingently rule-conforming, but pattern-governed (Seibt, 1990: 114). Moreover, as already insinuated, we can imagine that our explanations may never be complete. This means that, although capable of manipulating a meta-linguistic discursive level that reflects upon the use of terms in a specific community, one may employ terms according to rules of use that one would not be able to spell out, either because one is not yet acquainted with those rules or because those rules remain illegible to them in the current arrangement of reality. And again, when talking about language use, I do not simply mean uttering words, but any kind of aware or unaware activity that contributes to identifying the use of a term, both said and thought-out-loud. How we orientate towards the world is both conceptually determined and inscribed in the world we inhabit at a level that is both conscious and non-conscious.

This last aspect means also that even if one could spell out individual rules in use in a community, the person would not necessarily be able to change their individual compliance to rules of use, in purely volitional terms. In philosophical parlance, Sellars would state that one is not, at the same time, the subject of ‘ought-to-do’ and ‘ought-to-be’ rules (1973b: 492-3). One may display a verbal behaviour that is responding to an ‘ought-to-do rule’; that is, a behaviour the function of which is to bring it about that other people are in a certain state at a certain point. However, one is not bringing that state upon oneself, neither in the case when one is already in that state—because then the force of the ‘ought-to-do’ rule would be nil—nor in the case when one is not already in that state—because one does not talk oneself into a certain recognitional state. Mere talking is not enough. Things ought to be set up for one to recognize them as such and such. If a rule is only a verbal injunction and does not fit within a relevant set up in material reality, one will not be able to follow it; that is, one will not be able to be in the required state. The elements that are meaningfully at work in the process of norm acquisition are states or, Sellars would say, acts, rather than thoughts per se, and utterances that, in coordination with set actions, contribute to orientating us in the world.

A first point to deduce from the above account of language acquisition is an articulation of the Kantian notion that the relation between rules of validity and how we go about the world is not temporal. For a pattern to be at work, it is necessary that something prompts us to act in a certain way as opposed to another. This may very well happen without us being aware of it. What does it mean to say that at some point we become aware of it and how does this relate to rules and their coming about? At a certain point, we phrase a rule in words and sentences, which in and of themselves are only sounds and phonemes, but with which we effectively ‘coordinate’ a whole set of activities—and, again, this coordination may imply also acts we are not yet aware of or never will be. Only at this point is a rule in place as a rule, meaning not only that we can phrase it, but that we are aware of the fact that it commits us to seeing the world through certain relations. The value of a term becomes apparent to us in the web of rules of use, within a certain social set up, because the other nodes of the web, so to say, have also become apparent. Similarly, the activities that corroborate these appearances of meanings fall into place in the eyes of the one who can finally explain to themselves the use of the notions ‘yellow’, ‘red’, ‘colour’, ‘pencil’. Obviously, there is no guarantee that the rules we manage to phrase and through which a pattern appears truly match already ongoing socially shaped patterns, and even less that they match real processes in nature.

Now, two key aspects of this first point are that: rules could be wiring our sensations in a different manner from what we have come to phrase, and that it is only when they do any wiring that a pattern of continuity, or a previously unaware commitment, appears to us as a distinct behaviour. All that is available to us in terms of norms of use is available only from the side of rules, already phrased or phraseable, which is another way to say that when we open our eyes to a normative commitment, that commitment is already taken, individually and collectively. Sellars would sustain that, while we can infer from rule to behaviour—indeed behaviour is correlated with rules (1992: 173)—we cannot infer from behaviour to rule. Behaviour would not appear as behaviour if a rule that captures it could not be phrased. By highlighting this first point, we can appreciate that rules do not pre-exist language use, nor can we be sure that they perfectly match continuities of which we are not yet aware or will never be. We can account for validity from within language, as ‘ought-to-be’ states are linguistic and not natural, and need not appeal to either abstract or mental entities that would supposedly hide behind language and be voiced out into the world through it.

The second point that could be drawn out of Sellars’ account of language acquisition complements the first one. It entails that much more than re-labelling a term, or spelling out a different rule of use for the same term, will be necessary to place anyone into a different state, or in the condition of responding to a different ‘ought-to-be’ injunction, when interacting with the same portion of reality. A whole re-wiring of historical pattern responses is necessary and with them of the way that the world is materially organised. While this consideration is entirely unsurprising, articulating it through the distinction between ‘ought-to-be’ and ‘ought-to-do’ equips us with analytic tools that connect it with philosophy of language, in the first instance, and with the epistemic dimension of the medium, in the second.

As said already in various iterations, the notion of ought-to-be can be seen as fleshing out the fact that we acquire the use of terms through language, but in such a manner that the process of acquisition itself operates as a ‘vector’, ‘coordinating’ much more than the learning of supposedly valid uses of terms. By much more, I mean that, first, abiding to an ‘ought-to-be’, being in the state suggested by a certain ‘ought-to-be’, comes with less-than-aware acts, which cut into reality much more deeply than what the notion of state suggests. Secondly, ‘ought-to-bes’ imply behaviours that we are eventually capable of describing as such and therefore they entail also all that could-be-recognized-as-behaviours and that we, individuals or social groups, currently cannot name. Thirdly, when we finally become capable of recognizing ‘ought-to-bes’ for what they are—when we become capable of responding also to ‘ought-to-dos’—they constitute only the linguistic way into anything that has been transmitted to us, but we may never be able to observe as behaviour. This is to say that there seems to be no reason why we would limit the extent to which our cognition is shaped from the outside to observable behaviour, which only exists for us as a correlate to rule-bound uses of language.

In light of the above, we can, on the one hand, affirm that, in order to say that a rule of validity holds sway, we need not commit to either abstract or mental entities. The structures of what becomes conception do not exist independently of language and language is not simply a vehicle to voice them out of our minds. The acquisition of linguistic behaviour puts these structures into place. Semantic meaning and intentionality are derivative of verbal behaviour. On the other hand, we can state that the relation between language and what we can now call conception—broadly speaking, our world-view including all that goes into it that we are not and may never be aware of in linguistic terms—is not one of identity. Sellars’ verbal behaviourism, by differentiating ‘ought-to-be’ and ‘ought-to-do’ rules, shows a gap between our interactions with the world, as significantly shaped from the outside through the funnel of language acquisition, and the fact that our capacity to identify behaviour in correlation with a linguistic rule is only a posteriori.

Tellingly, Seibt’s reading of Sellars, when recapitulating the above explanation, indexes this relation between language and conception to the notion of medium. ‘Language is not the expression but the medium of our conceptualization’ (Seibt, 1990: 119). This is to say that when conception is accounted for in terms that do not ontologise abstractions, a notion of medium is appealed to, one that implies that a medium and what it mediates are not separate, since a medium does determine what it mediates—i.e., a different medium would mean a different manifestation of what the mediated is—nor are they exactly the same—the medium and what it makes appear may well be distinct and different. I will come back to this second arm of the relation once again in the following section, when the notion of appearance will be more meaningfully tied to the one of medium. The stance I take is therefore both non-materialist and non-dualist with regards to the relation between medium and mediated.

By virtue of being appealed to so as to identify the necessary alternative relation between language and conception (neither separate nor really the same), when accounting for conception, the undeniably broad and minimal definition offered above places the notion of medium at the core of the functioning of rules of validity. These are not eternal (Sellars, 1991: 298-320) but they formally, and therefore always already, hold sway on our purview of the world, when we begin to know it. The account of language acquisition shows that it is necessary for language to be a medium in the above sense to account for the working of the normative character of the transcendental, qua conditions of possibility of experience, its distinction from and mode of structuring the access to the empirical level. In similar vein, as I am about to flesh out, the language-medium is pivotal to the articulation of epistemological and ontological domains, according to which the ontological can be talked about only within epistemic constraints, but also separately postulated as autonomously existing.

After acknowledging that a certain medium—language—is indispensable for setting these distinctions, we can see how conceptions interact with other media in their modes of determining what they mediate, making something appear, making it accessible. The articulation of epistemology and ontology that I will now look at enables us to recognize for every medium an epistemological and an ontological relation with what it mediates—both relations in turn mediated by language and what it mediates, conception—refining the one given above. We will be able to hypothesise, from an ontological perspective, that the medium and what it makes appear can be immanent, entirely different processes. While, from an epistemological perspective, we will be able to say that the mediated has no status without a medium, only through which it becomes accessible, although different media may make it appear in different guises. The medium and what it makes appear are immanent and distinct, while medium and mediated are epistemologically impossible to parse from each other.

Importantly, while the structuring of transcendental norms, which constitute the formal break between transcendental and empirical levels is here described in terms of linguistically-mediated conceptions, this does not exclude that, if a different understanding of what it is to be a subject, unbound from the present anthropology, its correlative characteristic—conception—and prevalent medium—language—was possible, it would warrant a description of the formal character of validity in terms of a different medium.

 

The language-medium, the appearance of thoughts and the articulation of epistemology and ontology

One of the better sources to explain the relation that Sellars establishes between language and conception is Empiricism and Philosophy of Mind (1997, originally published in 1956). The declared objective of the article is to kill a myth (that of the given) with another myth (that of Jones). In the interest of our present investigation, we can describe this text as offering a clever configuration of the relation between epistemology and ontology, elucidating the relationship between language, conception, and, a third element, thought. This will also help confirm the definition of medium given above and especially the part that denies an equation of language and conception and places language in relation to thought or, at the generalised level, medium and what it makes appear in the mediated.

First, a brief recap of the story. Jones is a brilliant member of a fictional ancestral community, the Ryleans (named in polemic with Ryle’s behaviourism, 1949), who, as an ancestral species to our own, constantly think out loud as though their behaviour was ‘threaded on a string of overt verbal behaviour episodes’ (1997: 102). At a certain point, Jones realises that he can guess correctly his fellow Ryleans’ verbal behaviours, their thinking-out-louds, and hypothesizes: a Rylean’s apparent behaviour and related thinking-out-loud is the culmination of an inner—in the sense of quiet and private, an individually-located state—episode of sort. Having noted this, Jones teaches to the rest of his community how to report on the private episodes that their thinking-out-louds are supposedly the culmination of, therefore no longer needing to think-out-loud. This way of reporting is based on the possibility of hypothesizing that, when one finds oneself in a certain circumstance, one’s inner episode (or we may say thought) will be the one that usually culminates with thinking-out-loud word ‘X’. In this way, the Ryleans come to hypothesize the relationships among terms and the effectiveness of a certain inner episode. They build conceptual nets, which are conceptual because they always organize terms in relation to each other, where each term is the culmination of a hypothetical inner episode. Eventually, the mechanism of observing a certain circumstance and effectively postulating the occurrence of a related inner episode becomes automatic: inner episodes are no longer postulated as that which might culminate in the utterance of a term (whether silently or out loud), the Ryleans (and we) simply observe them as though they were the term itself, only now in its conceptual dimension. The passage from postulation in language to observation, or as Sellars would say from a theoretical to an observational function of language (1997: 181), signals the adoption of language as ‘natural language’ for our Rylean ancestors; that is, language as we, their offspring, know it.

For Sellars, the myth of Jones opens up our use of language vis-à-vis conception and shows that conception does not need to pre-exist language but can still be differentiated from it. As already pointed out, we do not need a dualist ontology to explain conception: the inner quality of linguistic episodes indexes their private dimension, although nothing that is abstract. At the same time, the difficulty we experience in parsing language and conception, Sellars would say, is due to the thousands of metaphorical years that separate us from this mythical beginning of language, which, by necessity, will always be mythical. What Sellars suggests we do is grasp the normative dimension that holds sway over our doings, from within language, and separately formulate a parallel hypothesis of the existence of something still entirely worldly. When we have coordinated the hypothesized entity with the verbal behaviours we observe as its hypothetical culmination, we may decide to label it with the generic term ‘thought’.

In terms of positively describing the language-medium, we could say that language carries a synthetic function. When perceiving a triangle, having an act of representing a triangle consists in perceiving anything along the lines of what we would then describe, for instance, as a-triangle-face-up-inclined-by-60-degrees. Yet, the object of representing is simply a triangle. In virtue of the fact that the object of representing is a triangle, we then manage to state our a-triangle-face-up-inclinated-by-60-degrees-ly perceiving (Sellars, 1976: especially para. 41, 42, 48, ff.). We will always have to work our way from the side of language, since the side of actual acts of representing is barred. In this, the object of representing or perceptible object is neither a mental act of representing a triangle, nor the sum of point-of-viewish mental acts, each individually representing a triangle. The reason is rather obvious: the perceptible object is linguistic, meaning that it is not made of the same stuff of thoughts, whatever that may be. Talking about synthetic function of language and, by extension, of the medium, means talking about a necessary and ineliminable jump between the level of what appears in the process of mediation, with regards to which we can say little, if anything at all, in terms of ontology, and the level of the perceptible object, the medium. The only way we have to comprehend this jump is by rendering in the medium itself what we postulate as acts of representing: thoughts as concepts, the postulated entity as what appears in whatever the medium mediates. Put in different terms, we only get to know—in the epistemic sense, according to rules of validity—thoughts in language as conceptions; however, there is a disparity between language, conception and thoughts. The language-medium stretches over the jump between the ontological and the epistemological dimensions, not in order to overcome it or to reduce it by any means, but because nothing would be accessible to our perception otherwise; that is, without the analogical function that is built into our capacity to approach reality. A ‘location’ of articulation is necessary, something in which we postulate the presence of something else, the actual existence of which is a metaphysical problem with no immediate (in the sense of unmediated) answer.

With regards to the above, I cannot overstate that the order of explanation of the argument, here, is not neutral. As noted already, it is precisely because Jones eventually decides to postulate the existence of thoughts as theoretical entities that the linguistic rules we observe—the normative and contentful character of which we have already justified independently of notions of conception—finally assume a conceptual dimension. The latter is the dimension that we (as opposed to the Ryleans) already normally attribute to language, just in a non-inferential manner.

To understand the non-inferential character of this attribution, and begin to reason also in terms of different media than language, we can think through another well-worn Sellarsian example (again from Sellars, 1997): that of molecules and gases. The observation of gases and their varying behaviour has prompted the postulation of the existence of molecules. And, it is because we have hypothesized the existence of molecules that we have begun accounting for ‘gas behaviour’ as we currently do. While closer to the time of the initial hypothesis (or while an apprentice scientist learns theories describing the behaviour of gases), observation of gas behaviour came accompanied with constant inferences in terms of what molecules would supposedly ‘be doing’. Later on (and for an expert), inferring what is happening at the molecular level by applying the rules that relate it to the gas-level is no longer needed. A savvy scientist looks at gases changing colour and simply ‘sees’ or rather observes molecular activity. Language would stand to thoughts, which supposedly appear to us as conceptions, in a way similar to how gases stand to molecules, which appear to us as gases changing colours, expanding, dissipating, etc. The parallel drawn here is one of rapports, because, from the perspective of their ontological character, molecules and thoughts are clearly very different. This aspect falls beyond the domain of this article, but the question as to whether Sellars’ philosophical project entails a metaphysical aspect (while still denying any ontology to norms) is a valid and indeed very interesting one.

Let us return to the reference made at the beginning of this section to Sellars’ configuration of the relation between epistemology and ontology—the hallmark of Sellarsian ‘materialism’—in order to align this with the difference I have just rehearsed between the observational and theoretical functions of language (1962: para. 14, 37). We access reality from an epistemic level, which itself ought to account for the mechanisms through which it assigns validity. Within it, we operate according to an observational mode. This means that we do not constantly infer from rules of validity to behaviours and use of terms, we simply speak and act in the world. Indeed, rules of validity usually hold sway in an undisturbed manner. However, there is much more to our acting than the rules of validity that we necessarily and only formulate in and as language. We access also this ‘more’ through the epistemic level, but we do so by formulating a separate ontological hypothesis and postulating the actual existence of some other process (choosing the notion of process in order to keep our metaphysics to a minimum). In this context, we operate according to a theoretical mode, whereby we postulate and infer the presence of processes that we can deduce only on the basis of activities that we see at play in what we can already observe. By virtue of its postulated existence, the process we infer changes our experience of the observable entity: what it is for us. Eventually, the entities we used to postulate simply appear to us.

Going back to Seibt’s quote and according to the above reading, to say that language is the medium of conceptualization is to say that language is that in which we observe concepts (networked terms for which a rule of use is at work, regardless our being aware of it or not) as the manifestations of something else. To say that language is the medium of conceptualization means that language is that in which thoughts appear to us as concepts. Language has to be a medium in the sense described above, for which the mediated (conception in this case) is not autonomous from the medium (language) nor is it equal to it, in order for us to be able to both recognise the inevitably linguistic character of conception, from an epistemic perspective (observational mode), and to formulate a separate hypothesis (postulational mode) over what appears in linguistically-mediated conception.

As said above, if we generalise the notion of medium reached in the account of conception from language to any other media, once the distinction between transcendental and empirical has been set and our ways of speaking in epistemological and ontological terms have been mutually articulated, we can say that, from an ontological perspective, the medium and what it makes appear are distinct and immanent processes. From an epistemological perspective, medium and mediated are instead inseparable. The motion picture film-medium is part of the complex mediation system that makes film appear, as the standard-length movie or as a field of artistic production which involves a whole variety of professions, machineries, economic investments, etc. by mediating cinematic images. Cinematic images would have no status without the rashes of photograms, but ‘film’ is not the same as them. The typewriter used to mediate book publications in which romantic literature, for instance, appeared. Blockchain mediates cryptocurrencies in which alternatives to the prevalent banking system might appear. Electromagnetic waves mediate, among other things, visual fields making energy appear—and it is because a certain range of waves is part of a visible spectrum, that we are able to access the ranges which are not visible. Video-game protocols mediate singular virtual images which make appear multiple modes of a player’s eye-body coordination. Clearly, these examples present media which are all linguistically and conceptually pre-mediated, but a generic definition of medium, such as the one deduced, opens up questions as to whether this pre-mediation is necessary and what it would mean for a subject’s interactions with reality for it not to be.

The philosophical formulation of the notion of medium proposed above is obviously broad, but, if adopted, it enables us to account for different kinds of media, which can then be themselves divided into classes. Importantly, it shows the role of the medium in terms of instituting a transcendental barrier that formally sets our capacity to validate knowledge, within a certain medium with regards to what it mediates. In other words, a notion of medium deduced from Sellarsian philosophy participates in an account of the formal character of knowing and the power of systems of validity from within a theory of the medium. Extending the definition beyond the language-medium to other media we can interrogate the relation between media and linguistically-mediated conceptions, whether all media are first linguistically mediated, what it would take for that not to be the case, what that would mean for our understanding of the transcendental, but also (and crucially) whether we could still be talking about knowledge and what other, more generic type of interaction, based on which type of subjective self-positing, would be the case.

 

Conclusion

What fruits can the encounter between Sellarsian philosophy and the authors working in the field of media theory mentioned above bear? Following the track of Sellars’ work, the article places a notion of medium at the core of a theorization of the transcendental because said notion becomes key to account for the working of validity, the formally a priori character of whatever shapes our experiences. The admittedly broad Sellarsian notion of medium here advocated is therefore of interest as it provides a frame for research on how media condition experience in relation to conception and for questions concerning how media may condition experience separately from conception. At the same time, the article links the notion of medium to the one of postulated theoretical entities and their appearance and implicitly argues for the philosophical dignity of the notion of medium. Indeed, this research highlights the necessity of the notion of medium for the articulation of a brand of materialism that is neither naive with respect to the epistemic constraints binding any kind of subject, nor foregoes the task of grasping reality beyond subjective apprehensions of it and the place of thought within it.

 

References

Balibar, E. (2015) ‘Foucault’s Point of Heresy: ‘Quasi-Transcendentals’ and the Transdisciplinary Function of the Episteme’, Theory, Culture & Society, 32(5–6), pp 45–77.

Benjamin, W. (2004) ‘On Language as Such and on the Language of Man’. (E. Jephcott Trans.) In Selected Writing, Volume 1, 1913-1926. Cambridge: Harvard University Press (Original work written 1916)

Brassier, R. (2007) Nihil Unbound. Enlightenment and Extinction. London: Palgrave Macmillan.

Brassier, R. (2011) ‘The View from Nowhere’, Identities, Journal for Politics, Gender and Culture volume 8, (2), pp. 7-23.

Brassier, R. (2013) ‘That which is not: Philosophy as entwinement of truth and negativity’, Stasis volume 1, pp, 174-186.

Delaney, C. F., Loux, M. J., Gutting G. & Solomon W. D. (1977) The Synoptic Vision, Essays on the Philosophy of Wilfrid Sellars. Notre Dame, IN: University of Notre Dame Press.

Ernst, W. (2013) ‘From Media History to Zeitkritik’, Theory, Culture & Society 30(6), pp. 132-146.

Ernst, W. (2016) Chronopoetics. London, New York: Rowman and Littlefield.

Foucault, M. (2005) The Order of Things. (A. Sheridan, Trans.) London, New York: Routledge. (Original work published 1966)

Foucault, M. (2002) Archaeology of Knowledge. (A.M. Sheridan Smith, Trans.) London, New York: Routledge. (Original work published 1969)

Foucault, M. (2008) Introduction to Kant’s Anthropology. (R. Nigro & K. Briggs, Trans.). Los Angeles: Semiotexte. (Original work published 2008).

Guillory, J. (2010) ‘Genesis of the Media Concept’, Critical Inquiry, 36, pp. 331-362.

Krämer, S. (2015) Medium, Messenger, Transmission. An Approach to Media Philosophy. Translated by Hanns, A. Amsterdam: Amsterdam University Press.

Krämer, S and Bredekamp, H. (2013) ‘Culture, Technology, Cultural Techniques – Moving Beyond Text’, Theory, Culture & Society, 30(6), pp. 20-29.

Kittler, F. (1990) Discourse Networks, 1800/1900. (M. Metter & C. Cullens, Trans.). Stanford, CA: Stanford University Press. (Original work published 1985)

Kittler, F. (1997) ‘There is No Software’. In:  Literature, Media, Information Systems: Essays. Johnston, J. (ed). Amsterdam: Overseas Publishers Association, pp. 147–155.

Kittler, F. (1999) Gramophone, Film, Typewriter. (G. Wintrop-Young & M. Wutz, Trans.). Stanford, CA: Stanford University Press. (Original work published 1986)

Macho, T. (2003) Zeit und Zahl. Kalender- und Zeitrechnung als Kulturtechniken. In S. Krämer and H. Bredekamp (eds) Bild Schrift – Zahl. Munchen: Wilhelm Fink, pp. 179–192

Macho, T. (2013) ‘Second-Order Animals: Cultural Techniques of Identity and Identification’, Theory, Culture & Society 30(6), pp. 30-47.

O’Shea, J. (2007) Wilfrid Sellars. Naturalism with a Normative Turn. Cambridge, MA: Politi.

Rosenberg, J. (1998) ‘Wilfrid Sellars (1912-89)’, Routledge Encyclopaedia of Philosophy, 1998. Retrieved from https://www.rep.routledge.com/articles/biographical/‌sellars-‌wilfrid-stalker-1912-89/v-1/sections/ontological-perspectives (accessed 24 September 2020).

Rosenberg, J. (2007) Wilfrid Sellars, Fusing the Images. Oxford: Oxford University Press.

Ryle, G. (1949) The Concept of Mind. Chicago, IL: University of Chicago Press.

Sachs, C. B. (2014) Intentionality and the Myth of the Given: Between Pragmatism and Phenomenology. London: Pickering & Chatto.

Seibt, J. (1990) Properties as Processes: A Synoptic Study of Wilfrid Sellars’ Nominalism. Atascadero, CA: Ridgeview.

Sellars, W. (1949) ‘Language, rules and behaviour’. In: Hook, Sidney (ed.) John Dewey: Philosopher of Science and Freedom. New York: The Dial Press, 289-315. Retrieved from http://www.ditext.com/sellars/lrb.html (accessed 24 September 2020).

Sellars, W. (1953) ‘Inference and meaning’, Mind 62(247), pp. 313-338.

Sellars, W. (1962) ‘Philosophy and the Scientific Image of Man’. In: Colodny, R. G. (ed.) Frontiers of Science and Philosophy. Pittsburgh: University of Pittsburgh Press, pp. 35-78.  Retrieved from http://www.ditext.com/sellars/psim.html (accessed 24 September 2020).

Sellars, W. (1969) ‘Language as thought and communication’, Philosophy and Phenomenological Research 29(4), pp. 506-527.

Sellars, W. (1973a) ‘Actions and events’, Nous, 7, pp. 179-202. Retrieved from http://www.ditext.com/sellars/ae.html (accessed 24 September 2020).

Sellars, W. (1973b) ‘Reply to Marras’, Canadian Journal of Philosophy 2(4), pp. 485-493.

Sellars, W. (1974) ‘Meaning as functional classification’, Synthese 27(3/4), pp. 417-437.

Sellars, W. (1976) ‘Kant’s transcendental idealism’, Collections of Philosophy 6, pp. 165-181.

Sellars, W. (1978) ‘Behaviourism, language and meaning’. Revised version of a paper presented in the ‘Symposium on Behaviourism and Philosophy’ at the University of Missouri (Columbia). Retrieved from http://www.ditext.com/sellars/blm.html (accessed 24 September 2020).

Sellars, W. (1991) Science, Perception, Reality. Atascadero, CA: Ridgeview.

Sellars, W. (1992) Science and Metaphysics, Variations of Kantian Themes. Atascadero, CA: Ridgeview.

Sellars, W. (1996) Naturalism and Ontology. Atascadero, CA: Ridgeview, 1996.

Sellars, W. (1997) Empiricism and Philosophy of Mind. Cambridge, MA: Harvard University Press.

Siegert, B. (2013) ‘Cultural Techniques: Or the End of the Intellectual Postwar Era in German Media Theory’, Theory Culture & Society 30(6), pp. 48-65.

Skinner, B. F. (1938) The Behaviour of Organisms: An Experimental Analysis. New York: Appleton-Century.

Sutherland, T. and Patsoura, E. (2017) ‘Michel Foucault, Friedrich Kittler, and the interminable half-life of “so-called man”’, Angelaki 22(4), pp. 49-68, DOI: 10.1080/0969725X.2017.1406047.

Weber, S. (2008) Benjamin’s -abilities. Cambridge, MA: Harvard University Press.

Withrop-Young, G., Iurascu, I. and Parikka, J. eds. (2013) ‘Special Issue: Cultural Techniques’, Theory, Culture & Society 30(6).

 

Silvia Mollicchi holds a PhD from the University of Warwick, where she focused on the work of Wilfrid Sellars and its relation to materialism, as articulated in media theory and contemporary philosophies of nature. She currently works at the Ada Lovelace Institute, where she researches the social implications and epistemic dimensions of machine learning technologies in various contexts.

Email: silvia.mollicchi@gmail.com

 

The version of record is available here:

http://journalcontent.mediatheoryjournal.org/index.php/mt/article/view/127