Occluded Frames (Draft of Truth and Context, Chapter Two (2001))
Thus far we have isolated both the region of explananda and the kind of explanans belonging ontological interpretation…
1.1 The Liar’s Paradox and ‘performance-reference asymmetry.’
Consider the liar’s paradox:
(LP1) This locution is false.
What is it that renders this paradoxical? Self-reference is the obvious explanation. The locution asserts that its own assertion is false. But how is it that a locution can assert something of its own asserting? Well, the locution refers to its own referring. The locution, in other words, performs something. What it performs is its reference. What it refers to is its performance of this reference.
The ‘short-circuit,’ then, can be understood in more substantive terms than the infinite oscillation of truth-values. The locution performs its reference to its performance of its reference to its performance of its reference and so on, ad infinitum. This is why a ‘paradoxical air’ hangs about self-referential sentences period, why
(LP2) This locution is true
is in some respects as strange, if not more. It is as though the relation of performance and reference in such sentences were analogous to the relation of frame and framed. In the performance of the reference to the performance, the ‘performative frame’ is precisely what is ‘framed.’ Munchausen like, self-referential locutions hold themselves up by their own hair. Part of the problem with such locutions, the suggestion might be, is that they violate what might be called ‘performance-reference asymmetry,’ that is, they attempt to frame their own frame.
This is not to say that ‘truth’ and ‘falsehood’ do not play the decisive role in rendering (LP1) ‘paradoxical,’ only that, in this incarnation at least, the Liar’s Paradox depends upon their application in a ‘symmetrical context,’ a context where the performance of a reference is the very thing referred to. Self-referential locutions such as
(SR1) This locution is printed in black ink.
for instance, do not seem to pose any particular problem despite their ‘violation’ of performance-reference asymmetry. (SR1) is in fact printed in black ink.
But despite the apparent intelligibility of (SR1), I would like to suggest that it is in fact unintelligible, and that it is unintelligible precisely insofar as it violates performance-reference asymmetry. Performance-reference asymmetry, I will argue, is a necessary condition of linguistic intelligibility, and as such, it reveals something profound about who we are and the way we are related to both our language and our world.
1.2 Performance-reference asymmetry and ‘frame occlusion.’
In order to see this, however, we need to consider another puzzling self-referential locution:
(SR2) The word ‘future’ is now in the future of this sentence, although by now it is not.
Prima facie, this sentence might seem merely nonsensical, and perhaps it is, particularly given that it demands ‘now’ to be understood in some kind of ‘super-indexical’ sense, that is, as referring to the very instant of its utterance within the time taken to complete the sentence. What interests me, however, are the two manners in which (SR2) might be read. If one considers the locution after the fact, for instance, there is a sense in which in it is actually true. It really was the case that at the instant indicated by the first occurrence of ‘now’ the word ‘future’ was in the future of the locution. If, on the other hand, one considers the act of reading this sentence for the first time, then it is hard to see how it could possibly make sense. It is only with the completion of the sentence–that is, after the instants indicated by ‘now’ have come and gone–that we can understand what it means.
Now one might object that this is trivial in two senses. Of course a sentence must be completed before we can understand what it means, and of course it is only after the fact that we can say that something was true of the future relative to a certain instant. The puzzle, however, is that the after the fact reading is not a reading at all. If the after the fact ‘reading’ makes sense, it is only because it assumes that the two occurrences of ‘now’ now refer to past moments of time–which is not what the two occurrences refer to in the act of reading the sentence. The after the fact ‘reading,’ in other words, is simply a retrospective consideration of our reading, rather than a reading. But the locution, as has already been pointed out, is meaningless in the act of reading. The significance of the ‘super-indexicality’ of the two occurrences of ‘now’ can only make sense after the locution is completed. It is as though, to rephrase a remark James once cribbed from Kierkegaard, ‘we read forward and understand backward.’1 This locution, it would seem, cannot be ‘read forward’ in any way that makes sense, and yet makes sense when ‘understood backward.’ Somehow we are able to make sense of a sentence we cannot read.
Something strange and yet profoundly important is going on here. Even if the ‘super-indexicality’ of the two occurrences of ‘now’ is something of a semantic peculiarity, it is not as though there is no ‘now of reading’ for them to refer to. The problem is precisely that there is. (SR2) is problematic because it refers to the immediate now of its performance, not because there is an immediate now of its performance. It makes sense ‘after the fact’ only because, safely ensconced in a further occluded immediate now, we can assign asymmetrical references to the two occurrences of ‘now.’ Only when the super-indexical ‘nows’ are assigned references to past nows through what might be called ‘retrospective ascent,’ that is, through a further ‘occluded now’ which does not make reference to the time of its own performance, can we make sense of this sentence.
This is both a banal and a remarkable observation: banal because obviously the sentence must be performed over time in order to refer, and remarkable because that very performance, as a condition of successful reference, must go on ‘behind the scenes’ as it were. Given our analogy of ‘frames,’ this suggests that if a certain performance ‘frames’ a reference, then this frame cannot be referred to, cannot itself be framed. Performance-reference asymmetry, in other words, entails what might be called frame occlusion.
1.3 Frame occlusion as necessary condition of linguistic intelligibility.
(SR1) This locution is printed in black ink.
then, we can clearly see how it is this locution seems intelligible, even though it is not. (SR1) is not intelligible simply because it cannot be read. If we consider (SR1) in Fregean terms, for instance, we can see that the argument that completes the concept expression ‘( ) is printed in black ink’ actually includes the completed concept expression insofar as the completed sentence is the referent of the argument ‘this locution.’ What this means is that the concept expression must be completed before it can be completed. And this simply means the concept expression can never be completed.
The reason (SR1) seems to make sense, I think, is that, after attempting to read the sentence, we surreptitiously extract the predicate and reapply it to the sentence in order to artificially complete the concept expression. What makes sense to us, then, is not (SR1) but something like
(SR11) ‘This locution is printed in black ink’ is printed in black ink.
In other words, we do something very similar to what was observed in the retrospective consideration of (SR2): we consider the locution from the standpoint of a further occluded performative frame, and so transform the performance-reference symmetry that renders it unintelligible into an instance of performance-reference asymmetry.
In fact, all properly self-referential locutions, that is, locutions that violate performance-reference asymmetry, are unintelligible. Performance-reference asymmetry, in other words, is a necessary condition of linguistic intelligibility. Every ‘referring to,’ one might say, must be at once a ‘referring away’ in order to be intelligible as reference at all. Our performative frame must be occluded.
Thus, returning to
(LP1) This locution is false.
we can see that the sentence is simply nonsensical rather than paradoxical. Once again, the concept expression must be completed before it can be completed, which is impossible. It is worth noting, however, that we do not read this after the fact as something like
(LP11) ‘This locution is false’ is false.
that is, we do not artificially complete the concept expression by reapplying it the way we might with (SR1). (LP11) here is clearly false rather than paradoxical. Given that (LP1) is nonsense, it is neither true nor false, thus it is false that ‘This locution is false’ is false. The question of precisely what artifice we use to extract ‘paradoxical sense’ from (LP1) is an interesting one, but not immediately pertinent to our present goals.
The argument here is that performance-reference asymmetry, the occlusion of the performative frame of our locutions, is an inescapable condition of the intelligibility of those locutions. Strictly speaking, then, there is no such thing as ‘immediate self-reference.’ Locutions such (SR1) and (LP1) strike us as ‘self-referential’ precisely because they attempt to perform references to the immediate performance of that very reference. But since these references only ‘succeed’ when we retrospectively consider our initial reading, which is to say, transform these self-references into ‘other references,’ we can see that such self-references do not in fact succeed at all. Only from the standpoint of a subsequent occluded performative frame can we frame the performative frame of our initial reading of the sentence. We make the unintelligible intelligible by tacitly adopting, via ‘retrospective ascent,’ a further ‘occluded frame of reference.’
Of course much needs to be said regarding the possibility of ‘circular reference,’ personal self-reference, and so on. These will be considered in due course.* The important thing to understand at this point, however, is that linguistic performances cannot explicitly refer to themselves.
Consider, for instance, propositional attitude ascriptions such as
(PAA1) I believe that p.
Here the proposition p and its performative context (my believing) would seem to be included in the same locution without detriment. But is this the case? Certainly not. Actually the performative frame of this locution is something like
(PAA11) I am writing ‘I believe that p.’
which is itself framed by the further performative context
(PAA12) I am writing ‘I am writing ‘I believe that p.’’
and so on, ad infinitum. In every case, the act of writing (or the act of reading) provides the occluded frame for the locution. I am sitting before the computer, 8:34 PM, September 13th 1999, in Nashville, writing these very words. But even in the course of referring to this performative frame, moments have elapsed. Even though I can refer to the performative frame of my referring I cannot refer, due to performance-reference asymmetry, to the further performative frame of this reference. Performance-reference asymmetry demands that the performative standpoint from which I write, ‘I am writing ‘I believe that p,’’ be occluded. The performative frames of our linguistic performances, one might say, always outrun us.
Henceforth, when I refer to our ‘occluded frame,’ I am referring to precisely the immediate performative frame that seems to constitutively outrun us, that ‘stands behind’ or ‘comes before’ us, as it were, and can only be made explicit from the standpoint of a further equally occluded frame. When I refer to our occluded frame, in other words, I am referring to this very frame, the occluded context of this very moment of reading–the moment which, of course, has already outstripped us in the process of being referred to.
2 Frame occlusion as a fundamental structural feature of perspectives.
Self-referential locutions such as those considered above provide a relatively straightforward way to make the peculiar, but undeniably important, phenomenon of ‘frame occlusion’ explicit. The significance of frame occlusion, however, is not restricted to the intentionality of linguistic performances. Frame occlusion characterizes intentionality period. Frame occlusion, in other words, is a fundamental structural feature of our relation to world.
Consider the manner in which our visual awareness of objects in the world takes the form of figures bounded by a broader visual field. When we focus on a red Camaro, for instance, we see something bounded by a peripheral field containing other possible figures that can likewise be bounded by subsequent peripheral fields. The most striking feature of this field, however, is that it cannot itself be visually figured. No matter how quickly we jerk our heads, we can no more ‘see the edge’ of our peripheral field than we can see the back of our heads. Our periphery, rather, trails away into an enigmatic ‘bounded edgelessness.’
Since figuration is only possible within our peripheral field, whatever limits that field cannot itself enter visual awareness as another focal figure. Our visual field, one might say, is the constitutive frame of the various figures framed in visual awareness. Hence, whatever frames our visual field, it seems, is necessarily occluded from our visual awareness as a condition of our visual awareness.
Both linguistic and visual performances, both ‘saying’ and ‘seeing,’ then, are characterized by frame occlusion to the extent that they both exhibit the same asymmetrical structure. In saying something we are directed away from our saying to what is said as a condition of saying, and likewise, in seeing we are directed away from our seeing to what is seen as a condition of seeing.
What is shared by both saying and seeing is this directedness, or as it is more commonly referred to, ‘intentionality.’ In this sense, the parallel suggested between linguistic and visual frame occlusion amounts to little more than the uncontroversial claim that both saying and seeing, as ‘modes of consciousness,’ possess similar intentional structures. What gives this parallel added bite, however, is the suggestion that what saying and seeing share is not merely ‘directedness,’ but directedness toward and away–which is to say, frame occlusion.
Frame occlusion belongs to all those ‘modes of consciousness’ that exhibit intentionality. With reference to other modes of perception, for instance, touch relates us to what is touched, smell to what is smelled, hearing to what is heard, and taste to what is tasted. In each case the perceptual mode is itself ‘transparent’: just as we cannot ‘see our seeing,’ we cannot touch our touching, hear our hearing, and so on. Each of these perceptual modalities, one might say, occlusively frames our relation to the world the before us. This seems to apply to the emotive modes of our consciousness as well. Desiring relates us to what is desired, fear to what is feared, and so on. Although it is possible, for instance, to desire desire and to fear fear, in both of these instances there is a distinction between desiring and fearing as intentional acts and desire and fear as the intentional objects of those acts, or put differently, between desire and fear as the frame of an intentional relation and desire and fear as they are framed by this intentional relation.
Of course, in each of these cases we implicitly ‘know’ that we are saying, seeing, desiring, and so forth, and in this sense, one might complain that it makes no sense to refer to these as ‘occluded frames.’ But the crucial point behind frame occlusion is not that we possess absolutely no ‘awareness’ of saying, seeing, desiring, and so on, but that this awareness fundamentally differs from our awareness of what is said, seen, desired, and so on. To paraphrase Wittgenstein once again: there is a way of grasping an intentional relation to the world that is not an intentional grasping of this relation, but which is exhibited in what we call ‘saying something,’ ‘seeing something,’ and ‘desiring something’ in actual cases. ‘Occluded frames’ simply furnish us with an onto-analogical means of making this distinction between immediate and mediate ‘graspings’ explicit. ‘Intentional grasping’ is like framing, the figuration of something within an unframed field.
An intentional relation to the world, in other words, looks something like
field figure field
which is to say, something like what we see all the time: figures framed by unframed fields. What this suggests is that intentionality is not a two-term external relation. Intentionality, in other words, does not look something like
O ————- O
where the unframed frame of our intentional relation to the world is framed (as ‘mind’) within a further occluded frame. What is lost in this depiction of intentionality is precisely the occluded frame structure of the intentionality it exhibits, a structure which looks something like
O ————- O
field figure field
Despite the fact that this figure/occluded-field structure is a condition of depicting intentionality as a two-term external relation, we overlook this structure in Fig. 2 simply because the field is occluded. Each of these figures, in fact, exhibit actual intentional relations in the process of analogically framing the structure of intentionality.
The concept of occluded frames, then, analogically makes explicit the general structure of intentionality as it is actually exhibited, not just here or there, but all the time. For us, intentionality is something like a transparent internal relation between an unframed field and a framed figure. The problem with Fig. 2 as a depiction of intentionality, then, is quite clear: it is nothing at all like intentionality as it is actually exhibited.
This is why concepts such as ‘mind,’ ‘consciousness,’ and the like, which we typically use to frame to the complex of intentionalities that constitute the ‘self,’ will be avoided in what follows. Such concepts tend to ‘overlook’ the occluded frame structure of intentionality, and so reduce frame/framed relation exhibited in actual cases of intentionality to an external relation between framed things. The only concept we possess, I think, that likewise refers to the complex of intentionalities constitutive of ‘self’ and yet preserves the occluded frame structure of intentionality is ‘perspective.’
2.1 The occluded frame structure of perspectives: a prelude.
Unlike ‘mind’ or ‘consciousness,’ it is very difficult to understand perspectives by analogy to things. The reason for this is that where the concepts of ‘mind’ and ‘consciousness’ take relations, there is a sense in which perspectives are inherently relational. A perspective, for instance, is always a perspective on something in the world from somewhere in the world. Unlike mind or consciousness, perspectives cannot be abstracted from their relations to the world because they simply are these relations to the world. One cannot, for instance, depict our perspective’s relation to the world as something like
O ————- O
because the world, although it certainly transcends our perspective, is nevertheless a necessary constituent of our perspective. A perspective on nothing is simply not a perspective. Our perspectival relation to the world, rather, always looks something like
perspective world perspective
which is to say, an internal relation between something framed and an occluded frame. The occluded frame structure of intentionality, in other words, characterizes our perspectives as well. The signature difference is that perspectives, like minds or consciousnesses, ‘possess’ a variety of intentional modes, such that it looks something like
desire world desire
which is to say, a complex of intentionalities likewise characterized by frame occlusion. Our perspective, one might say, consists of the ‘sum’ of our occlusively framed intentional relations to the world.
The key to securing comprehensive self-understanding, I hope to show, depends on a comprehensive understanding of the significance of frame occlusion as a structural feature of our actual perspectives on the world. In fact, the phenomenon of frame occlusion will come to occupy a position not unlike that occupied by the ‘ontological difference’ in Heidegger (which I take to be a misrecognition of frame occlusion). In what follows, I hope to show not only the pivotal importance of perspectival frame occlusion to self-understanding, but also the ways in which overlooking frame occlusion underwrites a plethora of philosophical self-misunderstandings.
3 The occluded frame structure of assertoric locutions.
Nowhere, I think, is the significance of frame occlusion, both with respect to its ability to provide an alternate self-understanding and to diagnose traditional misunderstandings, more evident than in the case of assertoric locutions. The occlusion of performative contexts, for instance, makes it strange to say that assertions refer to the world. If the performance of assertions is typically occluded, and this occlusion seems to be a condition of reference, then why do we say that assertoric locutions, the things that ‘disappear,’ refer to the world? In other words, is an assertion and its relation to the world something like this
O ————- O
where the assertion is a discrete entity (abstract or otherwise) externally related to a discrete feature of the world? Or is it, rather, something like this
[ O ]
assertion world assertion
where the assertion frames a certain perspective on the world? Of course, since the performance of an assertion is itself occluded, it should perhaps look like this
(assertion) world (assertion)
In the first characterization, an assertion is construed as something like a thing–be it a sentence, statement, or proposition–with a special relation to some feature of the world, a reference. In the second characterization, however, an assertion is construed something like a frame, as something that frames another’s perspective the world. In the third characterization, an assertion is construed as something like an occluded frame, as something that frames our perspectival relation to the world.
But an assertion is obviously a thing, is it not? As an inscription or an utterance this is certainly the case. But in of themselves marks and sounds are semantically inert. Marks or sounds only possess ‘semantic properties’ when taken up as competent performances within the context of a human practice. The semantic question, then, is not one of what kind of things assertions must be in order to possess the semantic properties they possess, but rather one of what assertoric locutions do, and how best to understand what they do. In terms of their semantic purport, in other words, assertions are ‘doings’ not ‘things.’
So what do assertions, as concrete linguistic performances, do? In the first characterization (Fig. 7), assertions simply refer to the world. This suggests that what assertions do, they do to the world. The problem, however, is that the world remains unscathed in the process of being referred to, and it accordingly becomes difficult to understand how assertoric linguistic performances ‘do’ anything at all. In the latter characterizations, however, this ‘doing’ finds a different direct object. Here assertoric locutions themselves do not refer to the world, rather, they occlusively refer us to the world.
When our interlocutors make assertions, it is not the world that is affected, it is us. We are convinced or unconvinced, amused or disgusted, but most importantly, we are directed to some feature of the world. Assertions refer us, one might say, in a manner analogous to the way a general practitioner might refer us to a specialist. Assertions redirect our attention to some feature of the world, and in doing so, they direct our attention away from the act of asserting itself. Assertoric locutions, the thesis is, occlusively frame aspects of our perspectival relation to the world.
Assertions, one might say, ‘move us’ to various ‘perspectival positions.’ They are merely ways to frame certain vantages or ‘angles’ on certain features of the world. Where walking from one part to the room to another affords different vantages on different objects in a room, uttering assertions transports us to different vantage points in ‘semantic space,’ as it were. When we comprehend an assertion, our perspective is literally ‘framed by a position.’
Prima facie, this alternate understanding is sure to sound strange, but it does possess a certain intuitive force. Prior to any instance of language use, we obviously possess perspectives that are, at least potentially, ‘on the world.’ And it is equally obvious that language use transforms our perspectives in some manner. Likewise, there is little doubt that assertoric locutions are characterized by performance-reference asymmetry–that what assertions do to us they do occlusively. The suggestion here is that assertoric linguistic performances enact momentary perspectives by ‘occlusively reframing’ our preexisting perspectives. Understood in this way, the function of language, in part at least, is to simply redirect our complex of preexisting intentional relations to the world, to transform our perspective on ‘this’ into a perspective on ‘that’ by occlusively redirecting our attention from ‘this’ to ‘that.’2
The idea, then, is that assertions frame particularities of our perspectival relation to the world in a manner analogous to the way that our position in a room frames our vantage on that room. This amounts to a literal construal of the metaphorics of ‘standpoint’ and ‘position’ that we commonly associate with assertoric discourse. Pretheoretically, we already implicitly understand assertoric locutions in this way. An understanding of the occluded frame structure of intentionality and perspective, I think, allows us to make this implicit understanding explicit. But the devil, as they say, is in the details. The only way to confirm this explicit onto-analogical understanding of assertions is to pursue and to refine it.
3.1 Diagnosis: Why assertoric locutions are mistaken for things.
The working thesis is that assertoric locutions occlusively frame perspectival relations to the world. When we frame assertions, however, look at them as objects possessing ‘semantic properties,’ these characteristics vanish. Not only do we artificially isolate assertions from the normative economies that frame them, as the contextualists complain, we also fundamentally misconstrue the relationality that belongs to them. Recall the temporal self-referential puzzle of (SR2), the fact that it was susceptible of two different readings. Even though the locution was initially senseless, we could make sense of it afterward. The locution, in other words, allowed for two different readings depending upon the standpoint one adopted. From the standpoint of the locution’s utterance or use, that is, while we were reading it for the first time, it made no sense because it violated performance-reference asymmetry by referring to the moment of its own performance. From the standpoint of the locution’s completion, on the other hand, we were able to make sense of it because ‘retrospective ascent’ allowed us to assign different moments to the instances of ‘now’ and so defuse the first reading’s violation of performance-reference asymmetry. When we construe assertions as discrete things we are adopting, I think, something akin to this latter standpoint.3 When we consider
O ————- O
for instance, we tend to overlook
[ O ————- O ]
that is, we tend to overlook the occluded frame through which these graphic symbols are related. We do so precisely because that frame is occluded. It always ‘stands behind’ or ‘comes before’ us, as it were. Given this, one might say that it never occurs to us that an assertion is actually
[ O ]
assertion world assertion
because in the act of asserting we see only
and in the wake of asserting, that is, when we return to the assertion from the standpoint of its completion, we see only
Assertions become something we look at rather than something we look through. And when we look at rather than through an assertion, it seems to gain an illusory independence from the act of asserting. And so, understanding that assertions have something to do with relations to the world, we make the almost inevitable mistake of construing those relations as something like
O ————- O
Assertions become discrete entities, linguistic or ideal objects, and as such externally related to the world they supposedly refer to. Consequently we find ourselves embroiled in the difficult business of explaining the properties of what is essentially a ‘though-at’ or ‘frame-framed’ relation in terms of an ‘at-at’ or ‘thing-thing’ relation–a task not unlike trying to understand how a pair of glasses will improve your vision without ever trying them on.
Not only, then, does the construal of assertions as occluded frames offer an intuitively plausible alternative understanding of assertions, it also explains why we mistake assertions for framed things possessing independent relations to the world. Assertions as frames are occluded as a condition of linguistically framing our perspective on some feature of the world, whereas assertions as framed from the standpoint of a subsequent occluded frame are not occluded. In redirecting our perspective on the world, assertions direct our perspective away from themselves.
As Aristotle writes, what is nearest to us is the most difficult to see.* This simply follows from the occluded frame structure of our perspectives.
3.2 Toward a perspectival theory of meaning.
Not surprisingly, this alternate construal of assertions as ‘occluded frames’ entails an alternate theory of meaning. On an occluded perspectival-frame account, I will argue, the actual meaning or content of any linguistic performance is nothing other than the concrete momentary perspective on the world it occlusively frames. In an important sense, we are the meaning of what we say and hear. And this, I hope to show, goes a long way toward explaining the difficulties we encounter whenever we attempt to isolate meaning as a kind of fact.
Part of the puzzle of linguistic meaning lies in the sheer number of dimensions it seems to exhibit. It seems clear, for instance, that understanding the meaning of a term has something to do with understanding how to properly use it. Meaning, in other words, possesses an important normative dimension. It also seems clear that understanding the meaning of a term has something to do with understanding what it stands for. Meaning possesses an important referential dimension. Likewise, it seems clear that meaning has something to with ‘speaker intention,’ that understanding the meaning of a locution depends upon understanding what a given speaker ‘means.’ Meaning possesses what might be called a private dimension. At the same time, however, it seems clear that meaning has something to do with communication. Meaning possesses a public dimension as well. It also seems clear that meanings possess associative relations to other meanings. Meaning, in other words, possesses a figurative dimension. And lastly, it seems clear meaning has something to do with truth, which is to say it possesses a logical and an epistemic dimension.
Linguistic meaning is at once normative, referential, private, public, figurative, logical, and epistemic. The challenge for any exhaustive theory of meaning is to somehow accommodate all these dimensions within a single explanatory framework.
4 Occluded frames and propositions: the problem of truth and context.
Thus far I have used the term ‘assertion’ to cover a number of different understandings of the semantic significance of assertoric locutions. Of these, the one understanding perhaps most unfairly characterized is that of the ‘proposition.’ *
Although it is clear that we can abstract sentences from utterances, and propositions from sentences, it is not at all clear that such abstractions are anything more than artifacts of our after the fact consideration of actual instances of language use. It is important, in other words, that we do not confuse our after the fact making explicit of different ‘levels’ of meaning with the actual existence of ‘entities,’ whether construed as abstract, concrete, or otherwise.
Nevertheless, there are some powerfully intuitive reasons for positing the existence of propositions aside from our ability to abstract different levels of ‘meaning invariance’ from utterances. Propositions are thought to capture that element of content or meaning that is relevant to truth-value and reasoning. Propositions, in other words, actually provide a means of tackling perhaps the most perplexing aspect of language use: the fact that, despite the profound contingency of our utterances, they nevertheless logically relate us to the world, which is to say, relate us to the world in a way that might be either true or false.
When one says, for instance,
(A1) The Camaro is red.
one is not implicitly saying
(A11) I believe that the Camaro is red.
even though such belief is certainly implied by the assertion. One is saying, rather, that the Camaro is red. We are directed away from the performative context, away from the utterance itself,4 and toward a certain feature of the world. We are not saying the Camaro looks red from here, but that the Camaro is red period, no matter what position we take in reference to it. Since the locution itself is a concrete artifact of some contingent context of use, belief, and the like, it is thought that, in order to logically relate us to the world across contexts, the locution must ‘express’ something that is somehow independent of contexts.
If some understanding of propositions seems indispensable to an adequate theory of meaning, Since an adequate theory must account for both the contingent and the logical aspects of meaning,.
The elaboration of an understanding of meaning that bridges this divide, the divide between truth and context, has been the‘holy grail’ of traditional semantics. The problem might be phrased as following: How can contingent linguistic performances in contingent contexts fund logical relations to the world? Thus far, however, propositions have contributed only to the formulation of the problem and nothing to its resolution. Positing two entities, the one concrete and contingent, the other abstract and logical, simply transforms the problem into one of relating these two entities.
4.1 The enigmatic relation between perspectives and truth.
But this problem takes a drastically different form when considered from the standpoint of frame occlusion. As has been suggested, locutions occlusively refer us to the world. The idea here is that locutions are best understood not by analogy to ‘things framed’ by our perspective, but rather by analogy to the ‘occluded frame’ of our perspective. Locutions ‘frame’ a momentary vantage on the world from a certain position in ‘semantic space.’ Given this, the question of how contingent locutions fund logical relations to the world is not the question of how two things, utterances and propositions, are related, but rather the question of how our linguistically framed perspectives, which are the result of concrete and contingent performances, can be ‘logical vantages,’ that is, true or false of the world.*
The primary difficulty this standpoint faces in answering this question, however, is that a perspective, by definition it seems, must be a contingent perspective. All perspectives on the world, linguistically framed or otherwise, are perspectives on the world from somewhere in the world, a ‘somewhere’ which we typically identify as the ‘here and now.’ What could be more contingent than a perspective perpetually confined to the here and now? And yet, as was suggested above, when we assert something as true, such as ‘The Camaro is red,’ we are asserting something that it is true period, and not something that is merely true ‘by our lights’ or ‘from our standpoint.’ In other words, we are asserting something that is true regardless of the contingent perspective we happen to occupy. This is why a ‘logical point-of-view’ is something of an oxymoron: there is a sense in which the truth or falsity of our assertions is something aperspectival.
In other words, the very notion of a perspectival understanding of language use would seem to rule out the possibility of accounting for the truth or falsity of our locutions. Given that perspectives seem inimical to truth, how could we possibly arrive at truth by starting with perspectives?
But despite this problematic outlook, I want to suggest that perspectives, once properly understood, actually offer a promising way to understand the ontological significance of truth and logic. For one, there is good reason to think that perspectives, the aforementioned difficulties notwithstanding, are nevertheless intimately connected with truth. With respect to (A1), for instance, what justifies the assertion as true, we would be inclined to think, is our visual perspective on a red Camaro. If, from subsequent vantages, the Camaro appears orange, we would be inclined to think (A1) is false. (A1) would no longer linguistically frame a vantage on how things are independent of our perspective, but rather how things were taken to be from that perspective. The Camaro is not in fact red; we merely believed it was read. There is little doubt, in other words, that perspectives bear in some important respect on the epistemic dimension of truth.
Another suggestive connection between perspectives and truth is found in metaphors such as ‘the view from nowhere,’ or the view ‘sub specie aeternitatis.’ How is it, one might ask, that these perspectival metaphors can capture any intuition regarding truth and objectivity? The first thing to note is that both a ‘view from nowhere’ and a ‘view from no time’ are every bit as oxymoronic as a ‘logical perspective.’ If a ‘view,’ or perspective, is always a view from ‘here and now,’ then the notion of a perspective that is neither ‘from here’ nor ‘from now’ would seem to be little more than an evocative absurdity. But this ‘absurdity,’ note, is simply the puzzle of our relation to truth: the fact that, despite our restriction to the contextual contingencies of the here and now, we can nevertheless lay claim to truths independent of the here and now. We remain confined to our view, but we speak ‘as though from nowhere.’
And this, strangely enough, suggests that truth is intimately related to the occlusive frame structure of perspectives. What allows us to understand the logical dimension of meaning without recourse to propositions, I would argue, is nothing other than frame occlusion.
5 A short history of frame occlusion.
In one guise or another, frame occlusion has played a pivotal role in philosophy at least as far back as Berkeley. BERKELEY, HUME, KANT, SCHOPENHAUER, NIETZSCHE
5.1 Berkeley and the occlusion of Spirit.
5.2 Hume and the occlusion of Self.
5.3 Kant and the Transcendental Subject.
5.4 The early Wittgenstein and the Metaphysical Subject.
But it is the early Wittgenstein whom, in his discussion of the ‘metaphysical subject’ in the Tractatus, glimpsed the profound relation between perspectival frame occlusion and logic. At issue in this discussion is a strange kind of limit, one which, like the limit of our visual field (an analogy Wittgenstein also uses), is constitutively occluded. Since the aim of the Tractatus is to sketch the limits of “the expression of thoughts” (*3), this immediately identifies these remarks as crucial to the Tractarian project as a whole. This occluded limit, Wittgenstein assumes, characterizes the ‘limit of our world.’ Since the world is the totality of facts, the limit of the world cannot itself be a fact belonging to the world. And since thoughts are logical pictures of facts, there is a sense in which the limit of our world cannot even be thought. At most, the limit of our world can only be exhibited in thought and language, for as soon as we try to represent this limit in thought or language, we transform it into a fact belonging to the world, which it is not. Hence the notorious metaphor of the ladder, the claim that the Tractatus constitutes a heuristic transgression of the limits of thought that must be dispensed with once understood, and hence the even more notorious self-negating imperative Wittgenstein uses to conclude the Tractatus: “What we cannot speak about we must pass over in silence.” (*74).
All of this can easily be expressed in the onto-analogical framework of occluded frames used here, albeit with far less tendentious consequences.* But what clinches the suspicion that Wittgenstein is actually discussing perspectival occluded frames are remarks such as 5.632: “The subject does not belong to the world: rather, it is a limit of the world.” The metaphysical subject, in other words, belongs to the occluded frame of the world. Unlike Kant, however, who thought this ‘occluded subject’ could be philosophically framed via transcendental deduction, Wittgenstein maintains that it cannot. The transcendental, according to Wittgenstein, can only be exhibited, never represented. Nevertheless, given the self-negating heuristic status of the Tractatus, he does make, directly or indirectly, a handful of provocative claims regarding this ‘liminal subject.’
For one, he claims this definition of the subject demonstrates that solipsism leads to realism. The truth of solipsism, he suggests in 5.62, is exhibited by “the fact that the limits of language (of that language which I alone understand) mean the limits of my world.” Once we see that the subject belongs to those limits rather than to the world, then we can see “that solipsism, when its implications are followed out strictly, coincides with pure realism.” He continues: “The self of solipsism shrinks to a point without extension, and there remains the reality co-ordinated with it.” (5.64/58) Now despite Wittgenstein’s elliptical presentation and aphoristic style, there is a sense in which he is actually following ‘strict implications,’ that is, given his understanding of the subject. As the necessarily occluded frame of the world, the metaphysical subject can represent only the world.
In other words, the truth of solipsism does not look like
[ O ]
self world self
where the world is framed by a framed frame, which is to say, by a subject. The threat of idealism looms large in this construal because the internality of figure/field relations can easily be mistaken, as it is in Berkeley, for dependancy relations. According to Wittgenstein, the truth of solipsism looks, rather, something like
where the frame is utterly occluded and only the world remains. In the absence of any frame there is no question of the internal frame-framed dependency relations characteristic of idealism.5 Hence we are left with ‘pure realism.’
Another indirect claim made by Wittgenstein regarding the metaphysical subject is that it coincides with the limits of logic. As he writes in 5.61: “Logic pervades the world: the limits of the world are also its limits.” Since Wittgenstein later asserts that logic is transcendental, this is misleading insofar as ‘pervades’ suggests logic is something in the world rather than the constitutive limit of our possible representation of the world.6 Since the metaphysical subject is the occluded limit of the world, and since the limits of the world are the limits of logic, then the metaphysical subject is also the occluded limit of logic. The suggestion here, then, is not only that logic is in some sense perspectival, but that, like realism, logic and truth have something to do with the occlusion of our perspectival frame.
5.5 The significance of these onto-analogical ‘near-misses.’
As we saw with Hume, there is a strange sense in which the frame of our perspective ‘stands nowhere.’ Kant, it was suggested, exploited this ‘nowhere’ in his bifurcation of the subject into ‘empirical’ and ‘transcendental’ aspects, and used transcendental deduction in order to bypass the empirical criterion and so provide an ontologically substantive account of the ‘transcendental subject.’ By working our way back from experience, he thought, we could secure a legitimate ‘synthetic a priori’ realm of metaphysical explananda. Thus imbued with content, Hume’s ‘nowhere,’ or frame occlusion, became something Kant could use, not only to explain the transcendental constitution of experience, but to secure the possibility of transcendental freedom and the objectivity of practical reason…
6 Truth and the ‘from nowhereness’ of the occluded frame structure of perspectives.
In some form or another, perspectival frame occlusion has played a profound role in the philosophical canon. But it is with Kant and Wittgenstein that one finds an explicit connection between frame occlusion and logic. Not only is this connection required in order complete the perspectival account of assertions considered above, it is absolutely essential to the development of an adequate perspectival account of meaning.7 As it stands, I can only give this account in pieces, slowly filling in the broader picture as more and more ontological insights become available. Only then will I be able to tell the full story of the relation between truth and context. At this juncture, the most I can do is make the relation between truth and the occlusion of our perspectival frame explicit.
The reason that vantages framed by assertoric linguistic performances like
(A1) The Camaro is red.
seem true or false irrespective of the contingent ‘from somewhere’ of their performance is simply due to the fact that this ‘from somewhere’ is occluded. Although our perspective is always ‘from somewhere’ in the world, this ‘from somewhere’ is always occluded. In other words, our perspective is always ‘from somewhere’ in such a way that it is also from nowhere. What this means is that our perspectives are at once ‘from somewhere’ and ‘from nowhere.’ And in a curious sense, they must be so. In the same way the frame of our visual field is nowhere to be found within our visual field, the immediate frame of our perspective on the world is nowhere to be found in the world. Certainly we can ‘reflect,’ frame this occluded frame, but only from the standpoint of some further immediate occluded frame.
Assertorically framed perspectives, the suggestion is, exploit this structural feature of our perspectives. They are quite literally ‘momentary views from nowhere,’ and as such they strike us as true or false independent of the contingent ‘from somewhere’ of the perspective they frame. As occluded, the contingent frame of our perspective vanishes much as Wittgenstein’s solipsist vanishes, leaving only that which is framed: how things are independent of our perspective on how things are.
The thesis, then, is this: assertorically framed perspectives possess logical truth-value by virtue of the ‘from nowhereness’ of frame occlusion.
6.1 Why assertoric locutions exhibit truth.
If this actually constitutes a more adequate ontological understanding of our logical relation to the world than that provided by propositions, then one might expect at least two things: 1) that it onto-analogically explain intuitions regarding logical truth that have hitherto been inexplicable; and 2) that it make some progress toward resolving those problems that suggest the inadequacy or the incompleteness of the propositional account of this relation.
With reference to the first expectation, this onto-analogical model actually explains why assertions exhibit truth-value. For instance, when we assert that the Camaro is red we are asserting that ‘The Camaro is red’ is true, and when we assert that ‘The Camaro is red’ is true, we are likewise asserting that ‘‘The Camaro is red’ is true’ is true, and so on. With each subsequent assertion, the truth-value of the preceding assertion is framed in a way that itself exhibits truth-value, an exhibition that can be subsequently framed by an assertion that likewise exhibits truth-value. What this suggests is that the truth-value of assertoric locutions always outruns the capacity of assertoric locutions to frame truth-value. Even if we were to continue this process of framing truth-values ad infinitum, we would still depend on the exhibited truth-value of the framing locution to frame the truth-value of our original assertoric locution.
Note the way in which this parallels our attempt to frame our occluded performative frame with respect to (PAA1) above. Our every attempt to frame the occluded frame of ‘I believe that p’ exhibited a further occluded performative frame, which, when likewise framed, simply exhibited yet a further occluded performative frame and so on. The reason for this, it was suggested, was that frame occlusion constituted the necessary condition of framing. Each attempt to perspectivally frame the immediate performative frame of our perspective required a further perspective whose performative frame was occluded. The reason, then, that assertoric locutions exhibit truth-value is that
What distinguishes this regress of occluded truth frames from the regress of occluded performative frames, however, is the sense in which each subsequent framing in the former simply repeats what was exhibited in the original assertoric locution. When one says, for instance
(A12) ‘‘‘‘The Camaro is red’ is true’ is true’ is true’ is true.
in a futile bid to outrun truth, there is a sense in which all one is saying is, ‘The Camaro is red.’ Whereas, when one says
(PAA13) I am writing, ‘I am writing, ‘I am writing, ‘I am writing, ‘The Camaro is red.’’’’
in a futile effort to outrun our occluded performative frame, one is obviously not simply saying, ‘The Camaro is red.’ In other words, there is a strange sense in which the occluded frame of performance in the regress of occluded performative frames ‘moves,’ where the occluded frame of truth in the regress of exhibited truth-values remains ‘motionless.’ In other words, where the stacking of intensional contexts in the former continually transforms meaning, the stacking of truth-predicates in the latter does not.
What this suggests is that the occluded truth frame, unlike occluded performative frames, cannot truly be framed. Unlike performances, we cannot actually frame truth, which is, strangely enough, exactly what our onto-analogical model suggests. Performances, which are always performances somewhere in the world, can always be framed precisely because they can be found somewhere. Truth, on the other hand, cannot be framed precisely because it cannot be found anywhere, and it cannot be found anywhere because it resides nowhere. Where the occluded ‘here and now’ of performance can always become a ‘there and then,’ the occluded ‘nowhere and no time’ of truth cannot. Where we can typically ‘step outside’ of a given performance and frame it from the standpoint of a further occluded performance, there is a sense in which we find ‘our backs against the wall’ with respect to truth. And this is simply due to the fact that there is ‘nowhere to go’ in ‘nowhere.’
I openly acknowledge how strange all this must sound, but is it really that peculiar that truth might be accessible to analogical understanding in this way? Certainly nothing bars the possibility of such understanding in principle. And there is, moreover, some sense in which these explanations capture ‘what truth is like.’ At the very least it conforms to our intuitions of truth as timeless and aperspectival. Nevertheless, the nagging suspicion remains that all this ‘explanation,’ and the understanding generated by it, simply must be ‘merely metaphoric.’ But if it is such, then it is in a way we don’t usually associate with poetic metaphors. Poetic metaphors do not exhibit the systematic comprehensiveness displayed here.
6.2 The ontological significance of the Equivalence Schema.
The above consideration of truth and the nowhereness of our occluded frame, for instance, seems to provide an onto-analogical explanation for what is at stake in the Equivalence Schema. Consider the disquotational version of the equivalence schema
(ES) ‘p’ is true if and only if p.
Most agree this captures something important about truth, though few agree what this ‘something’ is. What, for instance, is the difference between ‘p’ and p, such that the truth predicate is required to render them equivalent? Putatively, ‘p’ refers to an assertion that refers us to the world whereas p refers us to the world. The Equivalence Schema, in other words, might look something like
[ O ] is true if and only if O
One might then suggest that the truth predicate transforms
[ O ]
The truth-predicate transforms an assertion that is occlusively framed by our perspective from nowhere into one that occlusively frames our perspective from nowhere. What ‘‘p’ is true’ frames, in a sense, is the apparent framelessness of p, a framelessness or nowhereness that directly follows from the occluded frame structure of our perspectives. The ‘something important’ about truth expressed by the Equivalence Schema, then, is simply that truth pertains to the nowhereness of vantages framed by assertoric locutions.
6.3 Resolution of the ‘propositional attitude problem.’
Assertoric locutions frame vantages on the world that are apparently from nowhere in the world. This is why they appear to be true or false of the world irrespective of their position in the world. If this apparent nowhereness is a condition of their logical truth or falsity, then one might expect that framing these vantages relative to another frame would suspend their truth and falsity from the standpoint of our perspective. And indeed, this is precisely what happens.
On this account, for instance, a propositional attitude ascription such as
(PAA2) Aunt May believes that Peter Parker is good.
in fact looks something like
[ O ]
If this ‘occluded frame account’ possesses any truth, then it stands to reason that natural languages would possess the capacity of not only framing us in certain relations to the world, but of also framing frames as distinct features of the world. Propositional attitude ascriptions, I think, provide a striking a example of this capacity.
The problem with propositional attitude ascriptions, as traditionally conceived, is that ‘Peter Parker is good’ cannot be substituted salva veritate by say ‘Spiderman is good,’ even though ‘Spiderman’ refers to the same object as ‘Peter Parker,’ because Aunt May does not believe that Spiderman is good. The standing challenge of such sentences, or so we have assumed, is to find a way of preserving semantic compositionality without eliding the significance of propositional attitudes (particularly with respect to their usefulness in explaining behaviour), to render the assertion amenable to a truth-functional calculus without sacrificing the attitude.
But if logical truth arises as a result of way our linguistically mediated perspectives on the world are from nowhere, then it makes sense that the question of truth would ‘retreat’ in propositional attitude ascriptions, from the assertion that is framed to the assertion that frames–in other words, from the question of whether Peter Parker is a good man to the question of whether Aunt May believes that Peter Parker is good man.
For Aunt May the bare assertion that Peter Parker is a good man looks something like this
frame1 world frame1
where the assertion provides the occluded frame (frame1) of a certain perspective on certain individual in the world. The propositional attitude ascription that Aunt May believes that Peter Parker is a good man, on the other hand, might look something like
[ O ]
frame2 frame1 world frame1 frame2
where the propositional attitude ascription provides the occluded frame (frame2) of a certain relation to a certain feature of the world, namely, the frame of Aunt May’s vantage on the world (frame1). We are, in effect, framing a frame, or taking a vantage on a vantage. The question of truth retreats, as it were, from the first occluded frame (frame1) to the second (frame2), and so bears, not on the question of whether Peter Parker is a good man, but on the question of whether Aunt May believes that Peter Parker is a good man.
All ‘referential opacity’ in propositional attitude ascriptions consists in, then, are contexts where we are referred to framed assertions that in different contexts are able to frame us. Without any understanding of the nowhereness of the occluded frame structure that pertains to questions of truth, we have no way of understanding why assertions suddenly become referentially opaque in belief contexts when they seem characterized by referential transparency otherwise. Propositional attitude ascriptions thus become a ‘problem.’ We postulate entities such as ‘sense’ in an effort to devise a semantics that will accommodate the problem, or we attempt to explain the problem away through a consideration of the ‘ambiguity’ of intentional verbs, or, unwilling to relinquish theories of ‘direct reference,’ we merely define it away. On this account, however, propositional attitude ascriptions an expression of the …
6.5 Resolution of the ‘informativeness problem.’
[What follows has been appended from a previous draft]
If the meaning is given by the referent of each assertion, [the meaning is Venus. Since Venus is the common referent of the ‘morning star’ and the ‘evening star,’] then the question becomes one of how the assertion, ‘The morning-star is the evening-star,’ could be informative.
The problem, then, looks something like
[ ( O ] )
Morning-star perspective Venus Evening-star perspective
[where we have two different perspectives on the same thing, name Venus, without any knowledge that this is the case. For Frege and the early logical realists the question was one of retaining the centrality of reference while accomodating the ‘context sensitive’ element of meaning]. The problem, then, is the problem of the relation between the nowhereness and the somewhereness of our linguistic perspectives. How could a perspective be at once somewhere and nowhere? How can our perspectives be at once first person, third person, and third person omniscient?
Virtual transcendence as a necessary condition of communication. The ‘as-if’ nowhere.
7 Diagnosis: The broader philosophical significance of the occluded frame structure of perspectives.
With the equivalence schema we saw the ‘standpoint of truth’ advance from an assertion that is framed to an assertion that frames by virtue of the truth predicate. With propositional attitude ascriptions, on the other hand, we saw the standpoint of truth retreat from an assertion that frames to an assertion that frames that assertion by referring us to a belief context. Not only does this suggest that truth pertains to assertions insofar as they frame us, it also suggests an intimate relation between truth and perspectival frame occlusion.
[What follows has been appended from an earlier draft]
We all stand in the same nowhere. When another speaks, which is to say, linguistically frames our perspective on the world, they do so from the same nowhere. Assertoric locutions allow two perspectives to be one.
The picture I’m suggesting, then, looks something like the following. Minimally, the function of assertoric locutions is to transform
[ O ] ( O )
perspective A perspective B
into something like
[ ( O ] )
perspective A perspective B
which is to say, they allow two perspectives to be perspectives on the same thing. The function of discursive reason, on the other hand, is to transform this into
Consider, the present work. From my perspective of writing Truth and Context, I am in the possession of the truth, a truth which you, at least for the time being, do not share. From my perspective, then, things look like
( O )
world your perspective
[where you perspective on the world is skewed.] From an outside perspective, which is to say, from the vantage of someone who agrees with neither of us, the picture looks something like this
[ ( O ] )
my perspective world your perspective
But of course this is itself a perspective, and since I’m convinced I have my finger on the pulse of truth, the previous characterization is the proper one–from my perspective. The function of my arguments, then, is simply to transform [fig. 25] into [fig. 24], to continually linguistically reframe your vantage, linguistically reenact my own perspective on the world as yours, until you are convinced of the truth of Truth and Context. What I am trying to do, in other words, is to convince you to adopt my position.
[This actually highlights the primary shortcoming of contextualism:] it possesses no theoretical means to distinguish between [fig. 24] and [fig. 25], which is to say, no way to theoretically accommodate more. No way to understand context transcendence.
The signature difference between the position offered here and contextualism, then, is [that contextualism] remains perpetually stuck with
[ ( O ] )
Jerry Falwell Charles Darwin
and cannot theoretically account for the possibility of
( [ O ] )
C h a r l e s D a r w i n
which is to say, it cannot account for the possibility of more comprehensive perspectives. This is simply because the game rules out a theoretical account of more–of context transcendence. We need a way to arbitrate between ‘from somewheres’ in order to do this. Since contextualism can only offer an endless regress of exclusive somewheres (the veil of performances), it cannot do this. Somewhere’s must be able to transcend one another; they must include one another. The fact that they do do this is secured by my platitude [that a perspective is always a perspective on something]. (The question is not one of whether there is objectivity, but rather one of how we might account for objectivity.)