It's been a little while since I blogged on this topic: the last posting concerning it looked at the fourth section of the first chapter of Sellars' book Science and Metaphysics. That section uncovered the way in which Sellars' realist conception was distinct from that of contemporary physicalists and opened the way for delimitation of more characteristics of sense impressions. In this posting I want to look at the fifth section of the first chapter in which Sellars does not specifically refer to Kant.
This section opens with Sellars pointing out that, on the view set out in the fourth section, we can begin to understand the general temptation that exists to assimilate sense impressions to minimal conceptual representations due to the fact that the sense impression of something does not possess the typical qualities of that thing any more than a conceptual representation does. Having opened with this point Sellars rapidly proceeds to a distinct second one. This concerns the fact that "sense impressions" have only an analogical relationship to perceptible attributes of physical objects.
This point about the analogical relationship of the impression to the perceptible attribute is discussed in terms of the point that to speak of the impression "of" a red rectangle when the impression itself is neither red nor rectangular involves using the terms "red" and "rectangle" in an odd way. Sellars terms this use "adjectival". The point that he wishes to get at with this apparently purely verbal reference is that the general acceptance of the notion of "sense impressions" is as part of a causal story and in this causal story the sensory impression is taken to be a non-conceptual state which has a red and rectangular physical object as its standard cause. Sellars objects to this rendering of the sense of the "sense impression" as this requires it to be understood as a definite description through causal connections rather than as an analogical use of predicates and leads in the direction, once again, of "crude" (as Sellars now terms it) physicalism.
The third point that Sellars makes builds on this second one. Now Sellars points out that the causal story is also one that incorporates a kind of typical reference (what kind of cause could be expected) but that this is also odd in the case under consideration. After all, if the suggestion is that the sense impression of the red rectangle is, typically, produced by a physical surface that is red and rectangular then we reduce the description of the impression to a discussion of what it is that has produced it and, further, do so in a way that assumes that a certain feature of the physical surface can be basically identified with the sense impression (an odd assumption in many respects).
Sellars' fourth point is that when we say we have a conception "of" a red rectangle we are not making the same kind of reference to "physical object predicates" as we are when we speak about the sense impression "of" a red rectangle. This is followed by the fifth point that the temptation we are under when we conflate sense impressions with minimal conceptual episodes (something Sellars has consistently argued against in this chapter) is one relating to the impressions as if they were conceptual items of our "inner speech".
The final point of the fifth section concerns the fact that what is in common between "impression of" and "conception of" is that both are logically intensional as neither can lead to a direct inference that there are any objects of the sort "referred to" around. This intensional equivalence does not lead to an intentional connection as only the conception has intentionality. (This last point leads already to considerations that Sellars will take further in the second chapter of Science and Metaphysics.)
The last posting on this topic considered the third movement of the first chapter of Sellars' book Science and Metaphysics and concluded with a comparison of Sellars' view of Kantian intuition and that of Salomon Maimon. In this posting I am going to consider the fourth movement of the first chapter.
The fourth movement opens with a series of questions concerning the status of "sense impressions" considered as the content of "sheer receptivity". The basic question considers whether the "sense impressions" need to be considered as states of consciousness (albeit non-conceptual ones) that intervene between the causal impact of the external environment on the mind and the verbal expression indicating a referential relation to the environment. Rather than see the "sense impressions" this way it could be the case, as is suggested by contemporary physicalists, that all we need to discuss is physical states. In which case, the relationship of guidedness between the referential verbal report and the environment on which it reports would simply be carried out by the physical operations alone. Here, it is interesting to see how Sellars, whilst maintaining a framework of scientific realism, moves away from the standard responses of physicalists.
What the appeal to the notion of the "sense impression" involves is a kind of state that is neither conceptual nor physical and, in making this appeal, Sellars resists classical intellectualism on the one hand and brute physicalism on the other. (Here we see his intellectual kinship to Kant and why the Kantian reference is important to him.) The rationale for the "sense impressions" remains one of accounting for the appearance of conceptual representation within the general awareness of perceptions. Particularly, it is meant to help deal with what we might think of as "minimal conceptual representations" even though, as indicated in the third movement of the chapter, Sellars does not take it to be the case that all visual perception need involve these.
The basic question concerns one of a correlation between the physical objects on the one hand and the mental states that have a connection to them. This is expressed again, as was the case in the third movement, through an analytical response to the way that conceptual representations occur. So, the example is given that, under "normal" circumstances there are perceptions of a red and rectangular object in specific circumstances. The circumstances are when we are in some causal relationship to the object. Another situation in which the perception of this type occurs is an "abnormal" one where something else than the specific object is affecting us due to its possessing, in some sense, characteristics that are related to the object.
What Sellars wants to account for is not a set of behavioural characteristics such as are displayed when someone consistently shows they can distinguish between "red" and "blue" accurately. It is rather the ability to conceptualize the experience that is at issue. The basic problem is thus not one that concerns the second case mentioned in the last paragraph (the "abnormal" one) but the first case (the "normal" one). How is that the conceptual occurrence of the description of the datum is given when confronted with something that referentially matches said datum? In responding to this Sellars wishes to invoke the "sense impression" as what is at work in "guiding" the conscious state such that the "sense impression" immediately corresponds to the causal stimuli that brings it about. (This "immediate" element of the "sense impression" is also a connection of it to the Kantian sense of intuition.)
There has to be something about the "sense impression" that makes it, in a way that does not apply to the verbal conscious report, analogous to the causal stimuli that brings it about. Rather than invoke this notion Sellars mentions that we could have one of two alternatives that might seem to dispense with the need for it. We could, on the one hand, claim that the capacity in question, is innate. On the other hand, we could just deal with it, in quasi-Wittgensteinian terms, by speaking about induction into a language-game. Sellars does not attempt, however, to deal here with the rationalist appeal to innatism but simply leaves that suggestion aside.
Sellars' response to the quasi-Wittgensteinian view is more interesting. The point he makes here is that, in order for this induction into the language-game to work, there must already exist the possibility of some cues for connection between verbal reports and causal and physical stimuli which the story in question presupposes and cannot explain. The account is, in fact, part of a story of generational connection but not really one that makes clear how such connection can take place at all.
Going back to the question at issue Sellars distinguishes between the sense that there are impressions of red rectangles simpliciter and impressions that red rectangles are located in specific regions. The latter already requires something conceptual being added to the former and so the former is the real first point that needs accounting for. Similarly, in the first instance, we should not be thinking about the standard cause of the appearance of the impression as this will inevitably lead in the direction of Humean questions. The basic reason for that concerns the move, noted in the third movement, of minimizing the referential element when we become stuck on analysis of the occurrence in question in terms of a philosophical analysis that approximates to that of puzzlement of the ordinary perceiver.
I have been going through the first chapter of Wilfrid Sellars' Science and Metaphysics recently with the last posting on this topic discussing the second movement of the chapter and revealing how, in it, Sellars substantiated his conception of "sheer receptivity" and connected it to a revised account of "sense impressions". In this posting I will look at the third movement of the first chapter.
The third movement opens with what Sellars hopes is a relatively uncontentious claim to the effect that visual perception involves, in some sense, conceptual representations. If we do assume this to be so then it will follow that verbal reports made concerning what it is we are seeing will express a "conceptual episode" that contains the twin elements of perceptual capacity and a reference to how the general environment is causally affecting one. This point is then used to begin to explore the relationship between ordinary speech that refers to sense impressions and the philosophical analyses that have been undertaken of such. The contrast drawn is that, generally, people seem to have no compunction in referring to a general impression of a red book on a brown table but philosophical analysis of impressions rarely allows for the sense that what that phrase might report is a distinct sensory impression. The reason why philosophers appear to have been reluctant to analyse what is given in ordinary speech in a way that reflects the structure of that speech appears to have been that philosophical analysis of such impressions produces a thinner sense of what is given than ordinary speech appears to report.
However, there is a point at which philosophical analysis and ordinary speech get closer together and this is when ordinary speech reflects a problem in the speaker's response to the datum given. So, speakers become, for example, confused about what is given to them and then their speech reports become correspondingly thinner (and more like philosophical analysis). Sellars illustrates this with a progressive distinction between speakers concerning whether the datum they have given to them is "a red book over there" as opposed to "a rectangular physical object" or even it merely "looking like" there is a red book over there. The second claim would be a kind of minimal objective claim (much like philosophical analysis tends to give) whilst the last moves from a vocabulary that endorses something as given to one which places it in a merely "seeming" location. We can imagine the second claim relativised by also being put in a "seeming" manner.
If the second claim was so relativised then we have a kind of reflection on the statement in question. Now, in general situations the reason why this kind of reflection often appears required is, as first suggested, because of a kind of puzzlement about the datum given. It doesn't seem necessary to reflect otherwise. However, philosophical analysis has often operated as if the experience of the sense impression was typically like the one that ordinary speech only reflects in these cases of puzzlement. By contrast, what Sellars seems to aim at now is a distinction between the presence of minimal sense impressions and the necessity of minimal conceptual episodes like those given when we place the impression under the cover of a "seeming" notion.
However, rather than immediately directly arguing against the assumption that says a kind of minimal conceptual episode is required if there exist minimal sense impressions, Sellars first looks at what would have to be given in the minimal conceptual episode, if it existed. There would have to be, Sellars suggests, three distinct aspects of the visual perception. These would be: a) a purely physical aspect, however described; b) a primary mental aspect sufficient to produce the minimal conceptual episode; c) a "rich" conceptual episode that allows for reference to such things as "cabbages and kings and pigs in barnyards".
Sellars does not next instantly reject the division of the experience into the three elements in question. Nor does he reject the distinction between the "receptivity" of the sense and the "guidedness" of the sense by conceptual representation. The second is, in fact, taken to be required by Kantian analysis. For example, Kant refers to productive imagination "taking up" (A120) the manifold which implies, to Sellars, that the manifold is something independent. (See also, which Sellars does not cite, B145.)
However, whilst the latter point seems required for the Kantian analysis to work, this manifold should not be construed as belonging itself to the conceptual order but to be something non-conceptual. It is only if it is understood, in accord with the general pattern of Sellars' analysis, as so non-conceptual, that we have a notion that there is something "outside" the conceptual order, something that he takes to be a sense of "intuition", the sense at work in the uniqueness of perceptual judgment where the latter refers to some notion that there is something given, a particular this as such. This continued insistence on the understanding of intuition in this way is part of what makes Sellars' analysis so constantly reminiscent of that of Maimon whilst also distinguishing it from Maimon since the latter takes the consequence of this point to be that space and time are not really intuitions at all.
The general point that Sellars closes the third movement with is that the heterogeneity of the receptivity of sense (in the radical way required by the notion of "sheer receptivity") is what would be needed for Kant to escape the dialectic of German idealism that leads to Hegel. Without this heterogeneity we are well on the way to such idealism and the movement towards it seems, indeed, part of the effect of Maimon's analysis.
In my last posting on this topic I looked at the first movement of the first chapter of Wilfrid Sellars' book Science and Metaphysics whose argument culminated with a problem concerning the notion that space is the "form of outer sense". Today I want to reconstruct and assess some of the elements of the second movement of this first chapter.
Sellars opens the second movement by stating the up-shot of the first movement has been to suggest that representations of a manifold as a manifold, representations described by Kant as "intuitions", are a special class of representations of the understanding that belong to spontaneity, not to receptivity or at least not to "receptivity" in its "radical" or "sheer" sense. The "radical" or "sheer" sense of receptivity is, however, also now suggested to be a "postulate", albeit a transcendental one. The items which make up the manifold that belongs to the radical or sheer form of receptivity are, however, characterised by Sellars as "sense impressions" which ensures that what Sellars means by such impressions is somewhat different than the role they played in empiricist accounts.
Despite the fact that the place of "sense impressions" in Sellars' reconstruction of Kant is not equivalent to the role they play in empiricist accounts there is a consideration of how the role of this notion has often functioned within philosophy generally (and hence within empiricist accounts). The general account that follows focuses on the notion of an impression of a red rectangle which is analysed by Sellars as including three components. The first component is that what is meant by sensing this impression is that we are undergoing a conscious state. The second component concerns what is happening what this state is undergone which is that we are faced by a process in which this impression is, at least in "normal circumstances", brought about by physical objects that fit the appropriate description. The third component is that the impression in question represents the appropriate physical object.
Having analysed the experience in question through this discussion of its components Sellars goes on to discuss the understanding of the "state of consciousness". The phrase is itself described by him as "both ambiguous and obscure". The ambiguity is focused on first with the point initially made that such impressions, considered as states of consciousness, are often assimilated to bodily sensations and feelings. Sense impressions, in the broadest sense, are comprehended as "non-conceptual" and the impressions are understood as being not just "states" of consciousness but also as "objects" of it.
The question now arises as to whether the sense impressions are, as specific individuated "contents", themselves made objects of awareness (or, as Sellars puts it in more Kantian language, are "apperceived"). The point is made that once we grant the existence of such "impressions" it becomes almost impossible not to consider them as apperceived, particularly if we do not have care with our distinctions.
Cartesians, for example, have the possibility of saying that the specific, individuated, sense-impressions are objects of awareness but that they are things of which we are "inadequately" or only "obscurely" aware. Similar statements are made in a more recent vocabulary when philosophers suggest that we are not aware consciously of what the manifold of our senses "really" (physically, i.e., natural-scientifically) "represent". Sellars wishes to move away from such models precisely through his notion of "sheer receptivity" and its corresponding notion of "sense impression". So the point is that the classic uses of "sense impression" which develop this sense of "inadequacy" do so because they have not really distinguished the two senses of "receptivity".
In response, then, to the classic pictures of sense impressions we arrive now at a notion of them that allows for the possibility that such impressions are not specifically given to us as objects of awareness. In a sense, this fits, according to Sellars, with Kant's usage since he need not relate to any such manifold of "sheer receptivity" as something of which we are aware.
The second movement of the chapter appears, therefore, to consist mainly in developing and substantiating the concept of "sheer receptivity" and indicating the way in which it allows for a clarified sense of "sense impression" but Sellars, in this movement, does not yet fill out the place this will philosophically occupy for him in any serious detail and nor has he, as yet, returned fully to the way in which his renovation of the Kantian conception will both build on, and away from, Kant's own view.
Having begun to lay out both the views of Maimon and those of Sellars of the status of intuition I thought it possible to now pause and compare the two, at least in an initial way. Firstly, Maimon begins his discussion of space and time by arguing against a view of them as abstractions. In so doing he followed the logic of the arguments of the Metaphysical Exposition of space and time in the Transcendental Aesthetic. Sellars, by contrast, in his view of the dual status of intuition, comprehends the presence of intuitive individuals in cognition as a result of a form of abstraction. The difference between Maimon and Sellars here can be partially covered by attending to the point that Sellars' notion of abstraction is not from a passive manifold but rather from a synthesised one so the place of space and time in "sheer receptivity" would not be one that was a product of abstraction and the notion of abstraction he is working with is not in relation to the passive manifold but rather from the synthesised manifold.
When Maimon argues against conceiving of space and time as abstractions one of the reasons he presents is due to the claim that they are not parts of concepts of experience but are rather what gives unity to experience and, in being part of this unitary givenness, are related to the conditions of conceptuality as such. By contrast, Sellars lays stress on the immediacy of intuition which he connects to the model of intuition as a form of demonstrative "this". This difference can again be partially overcome by attending to the point that both Maimon and Sellars are stressing the sense in which space and time are conceptual though Sellars does not, just due to this, wish to abandon or leave as only "fictive" the status of space and time as intuitive.
The impact of Maimon's general account of space and time in the first chapter of his Essay appears to be that in experience there is no such thing as "pure intuition" but only what he terms "empirical intuition" which, I suggested, is as much as to say that we only experience "formal intuition" and not the "forms of intuition". I don't think there is a parallel to this argument in Sellars though his suggestion that there is a signal problem with thinking of space as the form of outer sense does begin to suggest a central difficulty with the very idea of "outer sense". Sellars' commitment to the notion of "sheer receptivity" seems to be a second element to his resistance to the reduction of space and time to being generically understood as conceptual with the the "intuitive" relegated to the status of illusion.
I have only as yet scratched the surface of Sellars' view and will return to it in more detail, also I will contrast the specific claims of Sellars with the interpretation of his claims by McDowell. The account so far given of space and time by Maimon has also, as yet, drawn only on the first chapter of the Essay and I will, in subsequent postings, move on to later chapters of it.
In the first posting I did on Maimon and Kant on intuition I referred to a connection between Maimon's response to the status of "intuition" and the views more recently advanced by Wilfrid Sellars and John McDowell. However, that posting grew so long that I was unable to come back to the views of Sellars and McDowell. In this posting I want to explore just some of the comments made by Wilfrid Sellars in the first chapter of his seminal work Science and Metaphysics: Variations on Kantian Themes.
The first chapter of Sellars' book opens by reporting Kant's heterogeneity thesis: namely, that there are two separate and distinct branches of cognition, namely, sensibility and understanding. Sellars relates this distinction directly to that between intuition and conceptuality and that between spontaneity and receptivity. Sellars aims to show that intuition has a double-aspect, in line with considerations we have been looking at in the last postings. The first argument to this effect concerns the way in which Kant describes concepts which is generally as a term for something general and not particular. On these grounds Sellars suggests a relationship between intuition and the presentation of individuals. However, the problem with such a view of intuition that Sellars considers, is to the effect that many particulars are clearly not of a sort to be presented as "intuitive". So, to incorporate this, Sellars points out that particulars are generally presented in such a way as to require mediation in their representation by concepts.
By contrast, with intuitions, Kant does specifically refer to the notion that there is something immediate about them. This claimed immediacy is something that is far from easy to understand and Sellars does not wish, at least at first, to endorse the conception that it should be understood as a causative claim (that is, that the presence of an intuition in a representation is brought about by something affecting the cogniser). Rather, Sellars' first suggestion is to the effect that the immediate particular that gets represented in an intuition is instead something that we should understand on the model of the "demonstrative this".
Now, if to represent an individual as a "this" is to be engaged immediately in an intuitive representation and yet the notion of such a "this" requires (as Hegel suggested in the "Introduction" to his Phenomenology) a basic conceptual relation, then this conceptual relation that is at work in the presentation of the "this" is one that belongs to the specific working of intuition. On these grounds, Sellars presents his first argument for thinking of intuition as, in some sense, conceptual. (Oddly, and more in keeping with Maimon, Sellars also refers to the way that Kant can include such a notion of an infinite intellect that is intuitive as an additional consideration for viewing intuitions as, in some way, capable of being comprehended conceptually though, unlike Maimon, Sellars does not develop this suspicion.)
Further considerations for approaching intuition in a manner which complicates the initial schematic representation of Kant's view quickly follow. Firstly, the representation of space is something that surely requires reference to understanding. This point is asserted and not yet argued. Secondly, the notion of the synthesis of imagination, implies a view of representing intuitions that involves some kind of meeting between spontaneity and receptivity and is certainly not a matter of what Sellars terms "sheer receptivity".
This conception of "sheer receptivity" is one that is not easy to work out the sense of in Sellars' use of it and I suspect I will need to say much more about it in future postings. Here it seems to be understood as the form of receptivity that involves no form of spontaneity. As such it appears that it would require a sense of "intuition" that is non-conceptual. So Sellars has now cashed out his view that there is a dual aspect to intuitions with one part being conceptual and one part not. The part which is conceptual would have to have a connection to judgment as is argued for in what I have elsewhere termed the "symmetry thesis" of the Metaphysical Deduction (A78-9/B103-4).
Sellars next compares the picture he has thus developed with the classic Aristotelian notion of abstraction pointing to the view that the judgment concerning an individual seems to arise from abstraction away from the individual presentation of something. In this case, unlike in the Aristotelian, the point would be that the individual intuitive presentation was itself already a result of active work and not something passively related to by the cognition.
However, despite the difference between the pictures, objections to the Aristotelian view surely resurface in slightly emended forms here. The suggestion that the presentation of a line, for example, emerges after the sense of the particular line has been given certainly seems implausible. Such a view certainly requires, at any rate, a strong commitment to the existence of a level of perception in which the determination of the "this" is working by means of something within the perception itself.
The connection of these reflections to Kant's argument arises from the point that "intuition" in Kant is, as pointed out by Henry Allison amongst others, at least distinguishable between the product of synthesis and a passive manifold towards which cognition is directed. This distinction is again related by Sellars to the difference between Kant's talk of a manifold of representations and the representation of a manifold. Whereas the latter involves synthesis the former may well not. (The explicit attention involved in the representation of a manifold as a manifold is something else entirely.)
Sellars now returns to his notion of "sheer receptivity" which appears to mark his view of what is at work in the manifold of representations being given to us. At this point Sellars assumes that this givenness by sheer receptivity is something that is simple and this suggestion appears to emerge from the sense that we are not here dealing with synthesis. Once synthesis emerges there is an evident complex but, in the absence of it, what need is there for a complex to be spoken of? That which appears to cognition without synthesis being needed ("sheer receptivity") would be a simple and so what is represented by means of it would also be simple. This equation is somewhat quick on Sellars' part since the means of representing may be simple without what is represented necessarily being so.
The case that begins now to be considered by Sellars is that of space which he argues cannot be given to "sheer receptivity" as a complex and, in arguing this, he relates the understanding of it to the notion that it is the form of "outer sense". Now, if space is part of sheer receptivity (intuitive in a non-conceptual sense) then, in being so part, it does not really fit with outer sense as such on Sellars' view. One reason is that such a simple form as would be given by space in this case is not fit to be the form of the representings of outer sense. If there is something given in outer sense then what is so given is surely not something that a simple mode of representing could capture and, because of this, it is also incapable of being what is represented by outer sense.
To be sure this does not directly tell against the view that space is the form of outer sense if we assume that space is an intuition of a conceptual sort but then it appears that space does not belong with "sheer receptivity" at all. This is the culmination of the first movement of Sellars' first chapter. Two questions arise from it: firstly, how does Sellars go on to account for intuition and, secondly, in what sense can the notion of "sheer receptivity" be shown to have a role in cognition? The latter question is one of some interest since it reflects a way of understanding the non-conceptual element of Kantian intuition, an element that we need some handle on.
Without yet moving beyond the discussion of Chapter 1 of Maimon's Essay I want to consider the account of space and time a bit further and, especially, the objections Maimon has to the notion Kant offers in the Transcendental Aesthetic that space and time are pure intuitions. This claim is one that the argument of Chapter 1 appears to be principally directed against (as is apparent from the closing sentence of the chapter). However when reading the chapter it takes some time to work out precisely what it is that Maimon is objecting to in Kant's characterisation of space and time as pure intuitions, not least because Maimon is happy to endorse the view that there are such things as pure time and pure space (the former being the object of arithmetic - number - and the latter being the object of geometry). These notions of pure time and pure space are distinguished by him from pure intuition as, Maimon claims, in the cases of pure time and pure space we have conceptual determinations. In support of this claim Maimon talks about time as a way of relating objects to each other and suggests that for the points of time to be distinguished we require a reference to something in the points of time. Here Maimon takes it that without some sense of substance and accident the points of time will not be separated.
This sense Maimon has of the way in which the distinct elements of space and time are distinguished for us in perception seems to be connected to his general problem with "pure intuition". Maimon seems to think of pure intuition on the model of absolute space and time and when he discusses the basis for the "fiction" that space and time are intuitive he argues that what happens in understanding space and time in this way is that something relative is taken to be absolute. The basis for this claim seems to be that if there is an homogeneous space then there are no parts sufficient to make difference available so that something heterogeneous is required. The heterogeneous that is required is in the matter and space/time are forms of relation between the heterogeneous matters but, if they were purely heterogeneous, there could again be no unity.
Maimon's arguments against pure intuition thus appear to be a form of Leibnizian argument against thinking of space/time as anything other than relations and an insistence that these relations only work to determine anything that is given independently of them which is why he appears to conclude Chapter 1 with the argument that space and time are only "empirical" intuitions or, as he also puts it there, are "predicates" of intuitions.
One question that emerges as fairly obvious from thinking further about this chapter is whether it is right, as Paul Franks suggested in his paper to the Maimon conference, to say that Maimon takes intuitions to be particulars. This seems correct in terms of Maimon's argument for thinking about the conditions of distinguishability of parts requiring specification and to point to why he has a problem with the view that there is a "pure" intuition but if that's right then Franks' point may need to be slightly refined. It is not that Maimon takes Kant's intuitions (i.e., his forms of intuition) as particulars but that he comprehends the representation that can be called "intuitive" (perhaps that of formal intuition?) to involve particulars in order for determination to take place. Though if that's right then perhaps Maimon's account of the "empirical" nature of intuitions can be re-phrased in Henry Allison's vocabulary as saying that his conception of intuition is that there are, in experience, only formal intuitions?
On beginning to read Maimon's Essay on Transcendental Philosophy the first question that arises concerns his treatment of "intuition", a treatment I think deserves extended contrast with the views of Wilfrid Sellars and John McDowell. Before going to the contrast, let's look first at some of the ways the first chapter of Maimon's Essay describes intuition. For the sake of today's posting I want to stick with the first chapter and not presuppose what is to come later.
Maimon's first point with regard to space and time fits with the argument of Kant's Transcendental Aesthetic. He opens by stating that space and time are not concepts abstracted from experience as they are not parts of concepts of experience but rather the unities through which the manifold "of experiential concepts" is gathered together. Whilst Maimon's manner of putting this is slightly different from Kant's it does not appear, at this point, significant in its difference since the argument appears to repeat the first argument of Kant's metaphysical exposition of the concepts of space and time in the Transcendental Aesthetic: A23/B37 and A30/B46.
Maimon also goes on to link this first point with a second which again echoes the argument of the Aesthetic. This second point is that the parts of space and time are not possible prior to the wholes of space and time, the parts are hence "in" space and time. This seems to conflate the 3rd and 4th of the second edition arguments from the metaphysical exposition (A24-5/B39-40 and A31-2/B47-8). This conflation erases the distinction between two separate arguments Kant gave for thinking of space and time as intuitions rather than concepts. Sticking with space, as is customary, Kant first gave the argument that space is not what he calls here a "general concept of relations of things in general" by claiming that the whole of space is prior to the representation of its parts. Subsequently he gave the separate argument that space is represented as an infinite given magnitude due to the fact that the parts are all contained within it, rather than, as with concepts, being instead thought "under" it. The distinction between the arguments is of some interest since the first was intended not merely to show that space should be seen intuitively rather than conceptually but also that the kind of concept it is not is one that describes relations of things in general. The second argument, by contrast, indicates the reason why the contrast of the first argument enables not merely a reference to the kind of whole that space is but also why this kind of whole is generically distinct from that given to concepts. So why does Maimon effectively conflate these two arguments?
The answer seems to be that Maimon wishes to point in a different direction to Kant when he lists these considerations of the way to view space and time. After mentioning these arguments Maimon immediately adds that he agrees with the Kantian claim that space and time are forms of our sensibility but then adds that they are grounded in the universal forms of our thought in general. In making this claim Maimon mentions that the condition of our thought is "unity in the manifold" and seems to be connecting the claim about space and time to the argument Kant will offer in the Transcendental Deduction concerning the transcendental unity of apperception. A reason for thinking that this connection is what Maimon has in mind can be found at A105 where Kant writes:
we think a triangle as an object, in that we are conscious of the combination of three straight lines according to a rule by which such an intuition can always be represented. This unity of rule determines all the manifold, and limits it to conditions which make unity of apperception possible. The concept of this unity is the representation of the object = x, which I think through the predicates, above mentioned, of a triangle.
As will become clear in subsequent postings, Maimon has a lot to say about the triangle, and, indeed, about geometry, a subject that will have to wait for another day (or, rather, many days!). The point is, however, that in referring to this view here, Maimon seems not to allow for an independent discussion of a priori intuition that Kant aims to carry out in the Aesthetic.
Maimon next goes on to mention the characteristic ways of understanding space and time, stating that space is the "being-apart" of objects (as follows from his first characterisation of space from the externality argument of the Aesthetic) whilst time is the preceding and succeeding of objects with respect to one another. It follows from this view of time that Maimon does not regard simultaneity as a determination of time, something that would at least cut against one reading of the 3rd Analogy, suggesting it needs to be read in terms of co-existence rather than simultaneity. However, more interestingly, Maimon is explicit in pointing out that space and time are conditions of each other in terms of representations, something Kant does indeed indicate but not always as explicitly as Maimon formulates it.
The next problem addressed concerns the view that space and time are conceptual as well as intuitive and it is here that Maimon's reflections relate to those of Sellars and McDowell. The claim that space and time are, in some sense, conceptual in addition to being intuitive is often thought (by Henry Allison, for example) to underly the distinction Kant makes between "forms of intuition" and "formal intuition" at B160. For reasons that would take too long to summarise in this posting I think this is a false view of the distinction Kant is drawing at B160 but that Allison thinks it plausible is sufficient to show that within contemporary Kant scholarship there is serious consideration given to the view that space and time are, in at least some sense, conceptual. Further, the argument, on-going since Kemp Smith's commentary on the Critique was first published, to the effect that the view of the Transcendental Analytic is not entirely of a piece with that of the Transcendental Aesthetic, further points in the direction of these types of considerations.
Maimon's way of articulating the view that space is a concept does not rely, at least in chapter 1 of the Essay, on either the considerations at work for Allison or those at work for Kemp Smith. Rather, Maimon simply goes back to the claim that conceptuality involves unity in the manifold and indicates that such manifold unity is also at work in the "concept" of space. The intuition of space is traced by Maimon back to the work of the imagination. In one sense, there is an obvious move here since intuition and imagination are closely connected for Kant. However, whereas the transcendental synthesis of imagination is, for Kant, the general basis of the possibility of experience (when combined with the transcendental unity of apperception), for Maimon, by contrast, to say that intuition and imagination are connected is to say there is something problematic in the intuitive sense of space. The intuitive sense of space is understood by Maimon as arising when something relative (such as place or movement) is taken to be absolute so the intuitive sense of space is an "imaginative" sense of it.
However, whilst this intuitive sense of space is therefore, in some sense inferior to the conceptual sense of space there is an important element of this intuitive sense of space that has to be conserved. This is that it is from the intuitive sense of space, according to Maimon, that "the objects of mathematics" arise. This claim introduces a further problem in Maimon's account since the possibility of the production of these "objects of mathematics" indicates a distinction within imagination itself as he makes clear:
The validity of the principles [Grundsatze] of these fictions is based only on the possibility of their production. For example, 'a triangle arises from three lines, of which two together are longer than the third'', or 'a figure cannot arise from two lines', and the like. In this case even the imagination (as the faculty of fictions, for determining objects a priori) serves the understanding. As soon as the understanding prescribes the rules for drawing a line between two points (that is, that it should be the shortest), the imagination draws a straight line to satisfy this demand. This faculty of fictions [Erdictungsvermogen] is, as it were, something intermediate between the imagination properly so-called and the understanding....
Here we find that the imagination that is "properly so called" is apparently distinct from the "faculty of fictions" as the latter is rather placed between imagination and understanding. The understanding itself is conceived here in orthodox Kantian fashion as the faculty of rules and is understood as active whilst the imagination is presented as partly passive and partly active. The passivity of imagination is the guidance of it by association (empirical or Humean imagination) whilst the activity of imagination involves the capacity to order. This active/passive combination is what is at work for "imagination proper" whilst the faculty of fictions is distinct from this as it is "completely spontaneous" (like understanding).
At this point things get further complicated. Maimon now presents his account of synthesis and describes three different forms of it. In one respect, again, this is orthodox since Kant did this also in the A-Deduction and section 26 of the B-Deduction still includes some sense of this. However, whereas Kant distinguishes between the syntheses of apprehension, reproduction and recognition, Maimon distinguishes rather between necessary synthesis, arbitrary synthesis and spontaneous synthesis. Necessary synthesis involves the unity and the manifold being, says Maimon, "given to the understanding, but not produced by it". This type of synthesis is at work, Maimon claims, when the real in sensation is given to us. This real is not produced by the understanding as it is sensible and it is necessary since sensation can be given no other way but understanding does not produce it. This is a suggestive response to the Anticipations of Perception.
The arbitrary synthesis, by contrast, is said to be produced by the understanding (hence not like the first synthesis just given to the understanding). However, whilst the understanding is the basis of this synthesis, this does not entail for Maimon that we have here an objective law. What does this synthesis connect to? Time and space as intuitively given and hence also the "objects of mathematics". As the first appeared to relate to the Anticipations of Perception so here there appears to be an echo of the Axioms of Intuition. Certainly the general claim that mathematics involves arbitrary production has a long ancestry in Kant's thought and I treated it at length in Chapter 6 of Kant's Transcendental Imagination. Maimon gives a lapidary presentation of his general claim here:
A determined (limited) space can be arbitrarily taken as a unit [Einheit] so that an arbitrary plurality [Vielheit] emerges out of it through the successive synthesis of such units with each other (this plurality is arbitrary as much in relation to the unity that is adopted, as with respect to the ever possible continuation of this synthesis).
However, whilst this again appears orthodox enough the way Maimon turns it shows an important twist since he goes on to claim that time and space as concepts contain "as differentials of the plurality, a necessary unity in the manifold" whereas Kant makes no reference to this conception, at least, not in this way.
Maimon also adds that the distinction between treating time and space as concepts and treating them as intuitions is important for thinking about the relation of time and space to each other. Considered just as concepts time and space are exclusive of each other whereas considered as intuitions they are not. The reason why time and space are not exclusive of each other when treated intuitively is then said to be that, as so treated, they are both extensive magnitudes, a claim that takes us straight to the Axioms of Intuition, just as I already claimed.
Finally, the spontaneous synthesis is a production of understanding from an objective ground but Maimon says no more about this in the first chapter. Rather, the discussion of the second synthesis steals the stage and from it Maimon appears to derive the surprising consequence that space and time, considered as "predicates" of intuition are only empirical and not pure though what effect this has on transcendental philosophy is not so obvious from the first chapter alone.
The considerations of this chapter are already very rich and I suggested, in opening, some connections to Sellars and McDowell but suspect I will have to leave the development of these to a further posting given the length this one has already run to. I will also treat later some considerations concerning how the views presented here relate to my previous account of Kant's view of intuition as developed in my book on imagination.
In my previous posting on this topic I began the process of examining how Rawls describes the distinction between ideal and non-ideal theory noting how the distinction arose initially in A Theory of Justice and how it was subsequently amplified in The Law of Peoples. I mentioned in this posting the latitude given to intuition by Rawls, a latitude that has something to do with Rawls complicated (and relatively under-explored) relationship with Henry Sigdwick. However, whilst this topic would itself be worth considerable discussion I propose now to look at the way in which this distinction is related by Rawls to his lexical ranking of principles.
Before looking at the question of lexical ranking, however, it is as well to remember the basis of non-ideal theory. If ideal theory is defined by Rawls as concerned with conditions of strict compliance then, naturally enough, non-ideal theory looks at non-compliance in certain respects but there are two kinds of non-compliance defined. One concerns "natural limitations and historical contingencies", the other how to "meet" injustice. Our duty is to remove, as far as possible, conditions of injustice but this is where intuition has a role. The lexical ordering of principles defines a way in which such intuitive judgment is guided. When this is done a further methodological device is invoked by Rawls in section 39 of Theory: that of the "representative equal citizen" (or, as he was once termed in British jurisprudence, "the man on the Clapham omnibus"). The reason for invocation of the notion of this representative is to apply the idea of the "common interest" where the common interest is a means of describing conditions that are to the advantage of all.
When this notion is applied to Rawls' general principles of justice some interesting results emerge with regard to how non-ideal theory is envisaged. So, for example, if we connect these notions to the general principle of liberty, then, states Rawls, there are ways that this principle can be legitimately restricted. So freedom of thought has to be exercised in a sense that is consistent with public order and this relationship is one that is part of what Rawls terms "the permanent conditions of human life" so this balance is one that is part of the area of non-ideal theory that concerns natural limitations. By contrast, if we look at other restrictions on liberty we can see how those connect with the other element of non-ideal theory. One way the principle of liberty gets restricted concerns the degree of tolerance given to those who lack tolerance themselves. This is part not of adjustment to natural limitations but instead to the principles for "meeting" injustice, or, as Rawls also describes this part, to the "partial compliance" part of non-ideal theory.
After outlining the relationship of non-ideal theory to the principle of liberty Rawls next turns to how it relates to the principle of equality. The restriction of application of the principle of equality occurs when unequal situations are allowed to arise and governed by some kind of rule. In some situations, for example, as envisaged in part by Kant in the Doctrine of Right, a division is allowed between active and passive citizens with the former having a larger role granted to them in public life than the latter. This inequality is not one that leads to systematic devaluation of those in the passive class since, as in Kant's own case, it can be fitted with a general argument against serfdom and slavery. Since the latter are ruled out on some appeal to the basic condition of liberty there is some sense in which this basic condition can be defended even though the application of it in principle permits unequal standing of citizens. So, in some sense, Kant gave priority to liberty over equality in relation to the standing of citizens (as well as in other respects) and Rawls agrees with this asserting the lexical priority of liberty over equality.
Rawls' own two principles are stated in section 11 of Theory and are worth citing:
First: each person is to have an equal right to the most extensive scheme of equal basic liberties compatible with a similar scheme of liberties for others. Second: social and economic inequalities are to be arranged so that they are both (a) reasonably expected to be to everyone's advantage, and (b) attached to positions and offices open to all.
When they are laid out it is evident that, whilst the principle of the lexical priority of liberty is similar in Theory to the one provided in the Doctrine of Right that the application of it is different. The second part of Rawls' second principle makes clear that the inequality permitted is not one of distinction between active and passive citizens since there should be a basis for anyone being able to hold any position. (It would be worth, on a different occasion, looking at the problems Kant would have with a similar declaration despite being often close to it.)
The second part of the second principle is often described as an "equal opportunity" principle whilst the first part is the Difference Principle. The second principle is lexically derivative of the first principle but, whilst both can be restricted, the ground of the restriction must be one that can be generally justified to and for those who are disadvantaged by its application. The point of including the discussion of the lexical priority of the principles is made clearer by Rawls in his late piece Justice as Fairness: A Restatement. In this late work Rawls suggests that the idea of the well-ordered society described in ideal theory should be able to provide guidelines for non-ideal theory including the problem of how to "deal with" ( or "meet" as he put in Theory) injustice. This requires, as we have uncovered, indicating the way in which injustice arises "naturally" in some sense as opposed to how it is part of real non-compliance. But this further needs to be understood by means of the lexical priority of the principles of justice since otherwise the degree of the real non-compliance would be entirely left to intuition. With the lexical priority added this intuitive element begins to be governed by a specific type of rule. In some subsequent postings I'll begin to probe the implications of these conceptions for how non-ideal theory gets laid out by Rawls.
