Acta Analytica

, Volume 27, Issue 4, pp 425–439

Time, Mode and Perceptual Content

Authors

    • Department of Social and Behavioural StudiesUniversity West
    • Department of Philosophy, Linguistics and Theory of ScienceUniversity of Gothenburg
Article

DOI: 10.1007/s12136-011-0134-0

Cite this article as:
Almäng, J. Acta Anal (2012) 27: 425. doi:10.1007/s12136-011-0134-0
  • 86 Views

Abstract

Francois Recanati has recently argued that each perceptual state has two distinct kinds of content, complete and explicit content. According to Recanati, the former is a function of the latter and the psychological mode of perception. Furthermore, he has argued that explicit content is temporally neutral and that time-consciousness is a feature of psychological mode. In this paper it is argued, pace Recanati, that explicit content is not temporally neutral. Recanati’s position is initially presented. Three desiderata for a theory of time-consciousness are subsequently introduced. It is then argued that a theory locating time-consciousness as a feature of psychological mode will fail to satisfy these desiderata. In the last section the intentionality of memories is discussed. Using the notion of shiftable indexical, it is argued that memories have the same explicit content as perceptions, but that they nevertheless can have different conditions of satisfaction since they are entertained in different modes.

Keywords

TimeContentPerceptionMemoryPsychological modeRecanati

1 Introduction

Most philosophers would agree that perceptual states in some loose sense of the word represent the environment as being at a specific (possibly extended) time. Two broad reasons can be given for this claim. First, it seems to be a phenomenological feature of perceptions that they present states as obtaining and events as occurring in a certain temporal order and with a certain duration. Secondly, temporal facts apparently enter into the conditions of satisfaction of a perceptual state. If I see a red ball right now my perception can only be veridical if there is a red ball in my environment at the present time. If there was a red ball in my presence an hour ago but which by now has been removed, the perception is not veridical.

While most philosophers would agree so far, there is widespread disagreement as to how perceptual states can represent temporal facts. One point of contention concerns whether or not time-consciousness is to be explicated at the level of psychological mode or at the level of intentional content. Recently François Recanati has defended the claim that it is to be explicated at the level of psychological mode. In this paper, it is argued, pace Recanati, that the time-consciousness of perceptual states is primarily a feature of intentional content rather than of psychological mode. Section 2 in what follows spells out Recanati’s theory. Section 3 argues that the tensed features of an act of perception are to be found at the level of intentional content. Section 4 presents an account of the intentionality of perceptions and memories that locates a temporal indexical in intentional content.

2 Recanati on Mode and Content

In his theory of intentionality, Searle (1983) makes a critical distinction between psychological mode and intentional content. According to him, the same intentional content can be entertained in various distinct psychological states. Thus, it is for example possible to see a red ball, to desire a red ball, think of a red ball, and so on. In these various cases the intentional contents can be identical, whereas the psychological modes are different. In the first case, the psychological mode is that of a visual perception, in the second it is that of a desire, and in the third it is that of a thought. Psychological mode is thus on Searle’s account the way a specific content is entertained by a mental state and by implication what makes the state into a state of a specific kind.

In Searle’s theory the psychological mode makes no contribution to the content of the state. It is at this point that he is criticised by Recanati. According to Recanati (2007: 130–135), we can distinguish between two kinds of content in perceptions. There is obviously the propositional or intentional content of the experience. This notion is equivalent to Searle’s notion of intentional content. In the following, we shall refer to it as the explicit content of the act. But explicit content differs according to Recanati from the truth-conditional or complete content of a perceptual experience. Let us assume that I am seeing a red flower. In this case, the explicit content of the experience would according to Recanati be “that there is a flower there”. On Searle’s account, however, the explicit content would also include a self-referential component. The explicit content would thus be tantamount to “that there is a flower there and that there is a flower there is causing this visual experience”.

Recanati does not deny the importance of the self-referential component. To the contrary, he agrees with Searle that it has an important function in the complete content of the perceptual experience. However, he denies that this component is a feature of the explicit content of the perception. Rather, he believes that it is a feature of the perceptual mode of entertaining a specific content. The reason is that in order for a mental state to be a perceptual state, it has to meet the self-referential condition: “For a representation that p to count as a perception that p, it must be the case that the representation is caused by the fact that p; but what is represented is only the fact that p” (Recanati 2007: 132).

The complete content of an act of perception is consequently constituted by the explicit content in conjunction with the contribution of the psychological mode. On Recanati’s account, the psychological mode makes the same contribution to the complete content whenever the explicit content is entertained in that specific mode. Indeed, the contribution to the complete content in an act of a specific kind is those features of the complete content that the act shares with other acts of the same kind. It is to be noted that this means that the phenomenology of an intentional state supervenes on the complete content and not merely on the explicit content. (Recanati 2007: 141)

The mode-content distinction forms the basis of Recanati’s reasoning when he argues that the time-consciousness of perceptual states is a feature of mode rather than of explicit content. Recanati notes that the critical distinction between perceptions and episodic memories of perceptual states concerns the temporal contents of the perceptual acts. When perceiving an event, the event is presented as occurring now. When remembering an event however, the event is presented as past. According to Recanati, we are inclined to think that when visually remembering a past event, the intentional act has the same explicit content as when we originally perceived the event: “The scene or event the memory is about is clearly the same as the scene or event the initiating perception is about (that is what makes memory)” (Recanati 2007: 137). In the following, we shall assume that this analysis is correct.

But even so acts of such different kinds as perceptions and memories can obviously not have the same truth-conditional or complete content. Perceptual states are veridical when the represented state of affairs is simultaneous with the act of perception. Memories however can only be veridical when what is represented lies in the past. Recanati believes that the solution to this apparent puzzle lies in assigning time-consciousness to the psychological mode of perception. The explicit content of perceptual states is temporally neutral. Yet perceptual acts are not temporally neutral since they have to be evaluated at a specific time, to wit, the present time. But this feature is something they have in virtue of being perceptual states, not in virtue of their explicit content. Here is Recanati:

The complete content of a perceptual state is analysed into (i) the explicit content of the state (the lekton) and (ii) a situation with respect to which that content is supposed to be evaluated. The complete content is distributed, and that means that what the situation component supplies need not be replicated in the lekton. Now the content of a perceptual experience is relative to the situation of perception. This relativity extends to time: the content of perception is temporally neutral, but it is evaluated with respect to the time of the perceptual experience. So the subject has, at t, a perceptual experience the content of which is the temporal proposition that there is a flower there, and that proposition is presented as true at t, the time of the present perceptual experience. (Recanati 2007: 141)

Consequently, time consciousness is according to Recanati to be explicated at the level of psychological mode rather than intentional content. In a similar vein, the explicit content of episodic memories is temporally neutral. Yet the complete content of the intentional act of visually remembering an event is not temporally neutral since the event is presented as past. But this is something the act has in virtue of being a memory, not in virtue of its explicit content. Consequently, the temporal consciousness involved in memories is to be explicated at the level of mode as well. It follows that a visual perception of an event and a visual memory of the same event have the same explicit content since that content is temporally neutral. Yet the acts have different complete or truth-conditional content since they are entertained in different modes (Recanati 2007: 130–135).

Recanati is not the only philosopher to attempt to locate time-consciousness in the psychological mode. Franz Brentano appears to have been attracted to this view as well (see Chisholm 1981). His approach is however radically different from Recanati’s approach since he attempts to account for time-consciousness by postulating the existence of several different but simultaneous acts. A closer resemblance to Recanati’s theory can however be found in the theory defended by David Woodruff Smith (1986, 1989). Smith has also criticised Searle for not recognising that there are two kinds of content and located time-consciousness in psychological mode, even though he does not exclusively locate it there (Smith 1989: 181–183). Smith’s primary interest is however not in time-consciousness, so he does not defend this position at any length. Consequently, we shall concentrate on Recanati’s theory in this paper.

3 Time Consciousness as a Feature of Content

Recanati’s argument seems to presuppose that in every perception, the time of the perception is represented in the complete content of the act in virtue of the content containing an indexical element referring to a “present” moment in time. Recanati is however less than clear as to whether this element refers to an extended, albeit very short, period of time, or whether it refers to a non-extended point in time. At this point, it becomes important to inquire into the nature of perceptual time-consciousness.

Any adequate theory of time-consciousness should satisfy three desiderata. It should be able to account for at least three features of perceptual experiences. The first desideratum is that it should be able to explain a feature of perceptual experiences that was much discussed under the term “the specious present” toward the end of the nineteenth century. It consists in the fact that perceptual experiences never seem to be momentary. Rather, we are always seemingly experiencing something as temporally extended. When listening to a tune of music such as “do-re-mi” you never seem to be merely conscious of a momentary tone such as “mi”. It is rather the case that you seem to be conscious of “do-re” when you hear “mi”. The three tones are perceived “together” in some sense of the word. It is not only the case that you first hear “do”, then hear “re” and then hear “mi”. When hearing “mi” you are still in some non-technical sense of the term conscious of “do” and “re”. Something similar seems to be true of visual perception. When seeing a ball move through a trajectory in space, you do not see the ball merely at successively different points in space. You see it as moving in space from one position to another, and you appear to be conscious of it as moving in a trajectory.

The notion specious present was originally introduced by William James and others in order to explain our capacity to retain past perceptual content in consciousness. One of the essential ideas behind this notion was clearly that what you perceive is never given in a mere unextended point in time, but rather as extended in time. Perceptual experiences essentially involve an experience of a certain temporal duration. Yet this experience is at the same time to be clearly distinguished from memory. It is not the case that you remember hearing “do-re” when hearing “mi”. Memories are a different kind of mental states than perceptions, and you need not be in any such state in order to experience temporal duration.

The second desideratum that I would like to highlight is originally due to Immanuel Kant. He noted that there is a difference between a succession of conscious states and a consciousness of a succession. The former does not entail the latter. Perceptual experiences are normally characterised not only by the specious present, but also by the specious present being temporally ordered such that it presents a succession of states to the perceiver. When you are hearing “mi”, you are not merely hearing it together with “do-re”, you are hearing it as succeeding “do-re” in time. In a similar vein, when seeing a ball move through a trajectory in space, you do not merely perceive the ball as being in different positions. You perceive the ball as occupying different positions in a temporal order.

Whereas the first feature of perceptual experiences highlights the fact that we always perceive objects or states of affairs as extended in time, the second feature highlights the fact that this duration is experienced as temporally ordered. So it is not the case that you first perceive “re” and then “mi” and then somehow judges that “mi” succeeds “re”. You literally perceive “mi” as succeeding “re”, and something similar obviously goes for visual perception. In short, objects or states of affairs are perceived as standing in temporal relations. We shall express this by saying that the object of a perception, whether it is a normal physical object, an event, or a state of affairs, is a temporal complex.

The third desideratum worth highlighting has been emphasised by Recanati himself. A type of a visual experience may be veridical in one situation and hallucinatory in another. Perceptual content must consequently contain an element referring to the time of the perception, or else they could not be assessed as to their veridicality. Any theory of time-consciousness must consequently explain how the truth conditional content can have this temporal feature.

At this point it is important to make a tristinction originally due to Alexius Meinong between time of act, content, and object (Meinong 1899: 245–247).1 The time of act is the time in which a person entertains a specific intentional content in an intentional act. The time of object is the time in which the object actually endures. The time of content is the time that the perceived object is presented as enduring in. It can easily be seen that time of content need not necessarily coincide with the time of the act. An intentional act can in principle be more or less momentary, yet present an object as enduring in time and undergoing certain changes. In this case the time of act does not coincide with the time of content. Following James, we can use the term “specious present” to refer to the temporal extension of the time of perceptual content.

In order to see how the tristinction works, consider for example the intentionality of beliefs. Trivially, a lot of our beliefs are about the past. I believe for example that the Second World War endured between 1939 and 1945. The time of the act is in this case whenever I have a belief to that effect. The time of the content, however, is the time between 1939 and 1945. Since the Second World War actually did endure between 1939 and 1945, the time of the object coincides with the time of the content. Whenever an intentional act is true or veridical, the time of content will coincide with the time of the object, though it need not coincide with the time of the act.

It is important to emphasise at this point that on Recanati’s account, time-consciousness is a function of the complete content, but not of the explicit content. The explicit content does not refer to any moment in time, and it is not in virtue of the explicit content that the act is evaluable at a specific point in time. This is clearly stated in the long passage from Perspectival Thought quoted above (Recanati 2007:141), and it is difficult to see how it could be different if time-consciousness is removed from explicit content. So according to Recanati, the explicit content is temporally neutral, and the complete content is tensed in virtue of the latter referring to the time of the act.

Now, if the above characterisations of perceptual experiences are correct, it is fair to say that the time of the content will always be temporally extended. However, this does not entail that the time of the act is temporally extended. Whether the time of the act is extended and consequently coincides with the time of the content is a major point of contention in the literature of the phenomenology of time, but we need not take a stand in that discussion here.2 As we shall see, both positions are difficult to combine with the doctrine that time-consciousness is located in psychological mode. It does not really help whether one conceives of the tensed indexical in the psychological mode as referring to an extended period of time or to a single point in time.

Let us initially assume that the tensed indexical contributed to the complete content by the psychological mode does not refer to an extended period of time but to a single point in time. If this is the case, it seems that time-consciousness cannot be a function merely of psychological mode. For, as we have seen, the perceptual object is always presented as a temporal complex. The complete content of the act presents a temporally enduring state of affairs. However, by assumption, the time being referred to by the psychological mode is momentary and not extended.

If we construe the indexical as merely referring to a single point in time, we end up with a theory that cannot satisfy a single one of our desiderata. Such a theory fails to account for the fact that the time of content is always extended. It also fails to account for the fact that we perceive events as preceding and succeeding each other. When hearing “mi”, you are hearing it as succeeding “do-re”. While you might possibly hear “mi” as sounding in the present and thus as simultaneous with the act, “do-re” is at the time of the act heard as sounding in the past and by implication as preceding the time being referred to by the psychological mode.

The basic trouble for Recanati is that the time of content for perceptions will always be extended. If the tensed indexical that is contributed by the psychological mode to the complete content refers to a momentary point in time, viz. the time of the act, then it cannot account for the fact that the complete content is extended and presents a temporal complex. But since these features are essential to the conditions of satisfaction of that very act, this theory cannot account for the veridicality conditions of the perceptual act either. In the case at hand, that part of content presenting “re” has its time of evaluation in the past. But if the content contributed by psychological mode determines the time of evaluation on its own and refers to an unextended now, then the complete content would have the wrong time of evaluation.

Perhaps it can at this point be retorted that since all perceptual experiences are experiences of something as temporally extended, the mode of perceiving would be to perceive something as present, but not as momentarily present, but as covering a short temporal interval that coincides with the specious present. This move can be made in either of two ways, either by assuming that the time of the act is momentary yet even so has a complete content reflexive to an extended “now-point” or by assuming that the time of the act is simultaneous with the time of the content and hence extended. In virtue of being extended, the time of the act would be reflexive to an extended time.

Considering that we never have an experience of a momentary non-extended perceptual state, I actually think that it is correct to argue that the tensed indexical in the psychological mode refers to an extended period of time. If the psychological mode explains our experience of being in a perceptual state, and if it is essential to this experience that we experience a stream of successive perceptions, then the tensed indexical in the psychological mode must refer to an extended now-point that includes the entire time of content. And this must be so no matter whether the time of the act is extended or not. But even if the tensed indexical in the psychological mode refers to an extended now-point, we cannot dispense with tense in explicit content.

A theory that posits an “extended” tensed indexical in the psychological mode seems to meet our first desideratum. The complete content would specify a short but extended period of time as the time of the act. Such a theory however fails to take into account the second and third desiderata mentioned above, viz. that we perceive temporal relations and that perceptual states should have the right conditions of satisfaction. Let us first consider the second desideratum. It is to be noted that the temporal relations perceived need not be identical from perception to perception. Even within the specious present we can discern different kinds of temporal order and different kinds of temporal distance between the relata that constitute the temporal complexes. Sometimes the specious present presents us with three events, sometimes with more than three events. The temporal distance between these events could of course vary as well.

When hearing “do-re-mi”, “mi” is presented as succeeding “do-re”. But this can certainly not be due to the psychological mode of perception. Even if the psychological mode refers to an extended period of time, it cannot present the explicit content of the act as ordered. But presenting “do-re-mi” as temporally ordered is what would be required of it if this line of reasoning were correct. What is needed is a kind of temporal operator that presents “mi” as succeeding “do-re”. Since the psychological mode will be the same whether we hear “do-re-mi” or “mi-do-re”, it cannot contribute such a feature to the complete content. If the tensed indexical is located in the psychological mode it could never give us the temporal order between the sounds we are hearing.

Consequently if the tensed features of the complete content are exhausted by the contribution of the psychological mode of a tensed content referring to an extended now-point, it is not possible to account for the perception of temporal complexes. In order to meet the second desideratum, the temporal content of perceptual experiences must also be located at the level of explicit content.

But perhaps, it could be argued, this criticism of Recanati is slightly unjust. He is clearly not trying to account for our first two desiderata for a theory of time-consciousness. He is above all interested in explaining the third desideratum, viz. the truth-conditional aspects of a theory of time-consciousness, and not how we perceive temporal relations or experience a specious present. So is not his theory of time-consciousness left unaffected by this criticism? Are we not dealing with two different kinds of time-consciousness, viz. one that explains our first two desiderata and a second that fixes the time of evaluation for the perception and thus explains why a type of perceptual content might be veridical in one situation and hallucinatory in another? This is however not the case. Whereas there might be more to the complete content with respect to tense than what is provided by the explicit content, the time of evaluation of the perceptual act must be a function of explicit content.

In order to see this, let us look a bit closer at the contribution of psychological mode to the complete content of the act with regard to tense. I have argued that the explicit content of perceptual experiences is not temporally neutral. However, this does not entail that the contribution to the complete content of the psychological mode is temporally neutral since we certainly have an experience to the effect that our perceptions occur in time. Consider first of all the perception of a red ball that is moving through the spatial positions p1–p3. The complete content of an act of perception of this movement would be something akin to

(i) The red ball is now moving through p3 after having just moved through p1 and p2and the red ball is causing this visual experience that I am having in this specious present.3

Now, I suggest that the italicised part of (i) is the contribution of the psychological mode to the complete content.

Note that on this account, the tensed indexical in the psychological mode refers not to a single point in time, but to an extended time that coincides with the time of the content. It is consequently a part of the psychological mode to experience the perception as present, but not as present in a non-extended now-point, since we do not experience a momentary point in time. Considering that that the psychological mode makes the same contribution to the complete content in all acts of perception, the tensed indexical must take as its value the time referred to in the explicit content. For if we assume that the time of the explicit content can vary with the situation at hand and is not constant, the time referred to in the psychological mode cannot be constant. Consequently, the only way that the contribution to the complete content by the psychological mode can be invariant from perception to perception is if the indexical takes as its value the temporal extension referred to in explicit content.

A strong case can now be made that the time of evaluation is a function of the time of the explicit content and that the psychological mode is rather irrelevant. In an act of perception we are presented with a temporal sequence of events. However, the tensed indexical that is contributed by the psychological mode cannot determine the temporal order in which the events are presented. For this indexical refers to the entire time of content without assigning a temporal structure to the events being perceived. Consider the event consisting of the ball moving through p3 and let us assume that this occurs at time t3 and that the ball moved through p2 at t2 and so on. The specious present thus spans time t1 to t3.

It is quite clear that the time of evaluation for that part of content presenting the ball as moving through p3 must be t3. But if we follow Recanati in assuming that the time of evaluation is fixed only by the psychological mode, we end up with a situation in which the time of evaluation is located sometime during the specious present, viz. between t1 and t3. The indexical contributed by the psychological mode is not specific enough to determine that the event occurs now, where this indexical refers to a single point in time. In a similar way, the time of evaluation for the part of content presenting the ball as moving through p2 must be t2. Once again, the time of evaluation cannot be determined by an indexical referring to the entire specious present, since that would give us too indeterminate a time of evaluation. That part would be determined as having the same extended time of evaluation as the other parts of content.

The “now” of the psychological mode is not singlehandedly able to give us a temporally distributed content, which presents the ball as moving through p3after having just moved through p1 and p2. It could only present these three events as occurring sometimes during the specious present but without specifying in which order. Consequently, the time of the evaluation must be a function of the explicit content in perception.

Perhaps an objection could here be raised. On the account that I am suggesting, would it not be the case that we experience the sequence of perceptions as unstructured, since the time of content refers to the entire specious present without assigning a specific order to the perceptions? But it is not true that the order would be unstructured. We experience the perception of seeing the ball move through p3 as succeeding the perception of seeing the ball move through p2 in virtue of the fact that we see one event as succeeding the other. The order in which we experience the perceptual states is determined by the order in which we perceive the events we are presented with through the perceptions. The perception of a sequence of events is temporally ordered through explicit content and the same order explains how we can experience a sequence of perceptions.

4 The Psychological Modes of Perception and Memory

One of Recanati’s main arguments for locating time-consciousness at the level of psychological mode rather than at the level of explicit content is that this ensures that memories and perceptions can have the same explicit content. Memories and perceptions have a temporally neutral explicit content on Recanati’s account. The difference in complete content is accounted for by the fact that time is a feature of mode: “Since the explicit content of the perception is temporally neutral, there is no objection to saying that it is preserved in memory” (Recanati 2007: 141).

Recanati is certainly correct in arguing that we have a strong intuition to the effect that memories and perceptions can in principle have the same kind of explicit content. Yet are we not by locating time-consciousness at the level of explicit content committed to denying this intuition? Recanati argues thus, but as we shall see, that is not the case.

Recanati considers two alternatives to his own account of memories: the “metarepresentational analysis” and the “conjunctive analysis”. The essence of the metarepresentational analysis is that it turns the original perception into a represented object. I agree with his objections to that theory and shall not discuss it further. It is however worth looking a bit closer at his analysis of the conjunctive analysis as outlined by Searle.

In Recanati’s reconstruction of Searle’s account of memories, the content of an act of remembrance has two distinct parts. On the one hand, there is a part that it has in common with the perception of which it is a memory. On the other hand, there is a part that is unique to the memory. The content is thus conjunctive: “The first conjunct represents a scene or event in the world; the second conjunct represents the subject’s past perceptual experience of that scene or event – hence the second conjunct is metarepresentational” (Recanati 2007: 138). Consequently, the explicit content (and according to Searle, this is the only content that there is) of an act of remembrance presents the original scene or event and in addition a remembrance of the perception of that scene or event.

Searle’s analysis of memories is not very clear to me, and I shall not try to defend his theory. I shall however defend the notion that a reference to the present tense features in the explicit content of a perceptual state and that that content in its entirety is replicated in remembrances. It is to be noted that these two claims are entailed by the claims that time-consciousness is at least partially to be explicated at the level of explicit content and that the explicit content of a perception can be replicated in an act of remembrance of that perception.

Recanati has two objections to Searle’s analysis. Both are relevant in the present context since they are directed to the idea that the present tense can feature in explicit content. The first objection is that Searle’s theory fails to take into account the difference between judgments based on perceptions and judgments based on memories. When making a judgment based on a present perceptual experience, you judge that what is perceptually presented is occurring now. When judging on the basis of a memory of that perception, however, you judge that the scene you perceived has occurred in the past (Recanati 2007: 139). The reasoning seems to be that if the explicit content of a perception is not temporally neutral but makes a reference to the present tense, and if the explicit content is retained when remembering a perception, it is inexplicable how we can justifiably judge that the event occurred in the past when remembering events originally presented in perception. Before returning to the first objection, we shall take a look at the second objection.

Recanati’s second objection is that it seems phenomenologically false to claim that the explicit content of an act of perception has a present tense and that this content is replicated in an act of remembrance. When the content of a perception is transformed into a memory, the present tense is subtracted from the complete content of the original perception and replaced with a past tense. There is a “feeling of presentness” accompanying every perception that is shifted to a “feeling of pastness” in an act of remembrance (Recanati 2007: 140). Recanati can make this move because he insists that phenomenology supervenes on the complete content rather than on explicit content. Consequently, the explicit content (which is tenseless) can remain the same even though the phenomenology of tense varies between different kinds of acts.

It seems to me however that Recanati is mistaken in both objections. It is possible to construe a theory that retains the present tense in the explicit content, yet which escapes Recanati’s objections. Recanati argues that there is a “feeling of pastness” accompanying every memory. But is this really true? I would suggest that it is far from obvious and that in so far as we have intuitions to that effect, they can be given a different explanation than that there is a “feeling of pastness” in the act of remembrance.

Consider first of all the very act of remembering having seen an event in the past. We are disposed to judge that this event has occurred in the past in virtue of the fact that we remember an event. We do not in addition need the memory to represent the event as past. All normal persons believe when they are remembering something that the events represented in memory have occurred in the past. It is an essential part of an act being a memory that the event represented did occur in the past, if it did occur at all. And normal cognizers know this in virtue of being familiar with the kind of mental state that an act of remembrance is. If we are presented with an event through a memory, we do not in addition need to represent the event as past in order to judge that the event occurred in the past.

A second consideration that should be borne in mind when discussing the intuitions of Recanati is that visual memories are in the stream of consciousness often embedded in non-visual propositional thought. So, for example, we might think, “the summer of 1994 was a summer with fantastic weather” and then immediately following on this thought is a train of visual memories of the sun setting across a lake and other environmental scenes. In this case the total stream of consciousness certainly has a feeling of pastness, but is it really true that the visual memories in that stream have a feeling of pastness? I would suggest that that is not the case.

Consider our explication of the content of a visual perception in (i). It is characterised by the fact that the perceiver seems to perceive the red ball while at the same time being, in some non-technical sense of the word, aware that the red ball is presented in the mode of perception. The act is further characterised by the fact that it contains a dual kind of time-consciousness. On the one hand, the content presents the ball as moving in the present and coming from the past. But on the other hand, the content also makes a reference to a perceptual experience of the ball as occurring in the present tense. Something similar, I would suggest, is involved in a remembrance of that perception.

When we do remember an event, we experience the act of remembrance as occurring now. The key difference between seeing an event and remembering seeing an event is not in the tensed indexical contributed by the psychological mode – in both cases there is a present tense. The difference is that in the first case we perceive the event and in the second case we remember having perceived the event. But both experiences are experienced as occurring now.

But on the account that I am defending, there is also a second temporal feature in every act of remembrance. And this is the tense of the explicit content that is replicated from the original perception. The explicit content does not merely contain an indexical referring to the present tense. If we are to account for the second feature of time-consciousness, degrees of the past tense seem to be involved as well. Otherwise we could not remember seeing (or hearing) a succession of events. But the present tense clearly features in the explicit content. Or at any rate this is how it appears to the present author. It is however notoriously difficult to agree on phenomenological intuitions. I shall thus leave that particular problem and proceed to argue that this account does not fall prey to Recanati’s first objection.

Based on these considerations, the following complete content of an act of visually remembering the ball moving suggests itself:

(ii) The red ball is now moving through p3 after having just moved through p1 and p2and I am now remembering my visual perception of this.

Once again, the contribution to the complete content by the psychological mode is italicised. It is to be noted that the contribution to the complete content of the psychological mode in (ii) would be rather complex. It would correctly explain why a memory is experienced as a present experience and our sense when remembering of having had a particular perceptual experience of an event as present. Yet this account does not make the experience into an intended object (Recanati correctly warns against such a move).

If we look a bit closer at the logical structure of (ii) with respect to tense we shall see that the complete content would look like this:

(iii) Presently I remember visually perceiving x as present.4

This however leaves us with an apparent problem. For whereas (iii) might do the phenomenology involved justice, it seems that we end up with conditions of satisfaction, which specify the time of evaluation for remembered events in the present. But that is clearly erroneous.

The solution to this problem lies in treating the tensed indexicals as shiftable indexicals.5 The context of evaluation for the tensed indexical in explicit content is shifted from the time of the act to a time in the past, in virtue of the explicit content being embedded in an act of memory. So whereas the explicit content still presents the person entertaining the content with an event in the present tense, the time of evaluation is nevertheless located in the past in virtue of the mode in which the content is entertained.

Philippe Schlenker, one of the originators (cf. Schlenker 2003 and Schlenker 2004) behind the idea of a shiftable indexical, uses the notion to clarify (amongst other things) what he calls “The Historical Present”. A narration occurring in the historical present is characterised by the author using the present tense when describing a sequence of historical events. Here is an example used by Schlenker himself:

Fifty-eight years ago to this day, on January 22, 1944, just as the Americans are about to invade Europe, the Germans attack Vercors. (Schlenker 2004:281)

In this case, the author’s use of the present tense clearly denotes an event that is past relative to the narration of the event. As Schlenker points out, the “result is to present the scene in a particularly vivid way, as if the narrator were observing it directly” (Schlenker 2004:297). The general idea is consequently that when one uses the historical present, one locates the point of view of the narration in the past, thus allowing the author to use the present tense in order to refer to past events.

In order to explicate how a present tense can refer to a past event, Schlenker makes a distinction between the context of utterance and the context of thought. The general idea is that in narrations using the historical present, the context of thought is the context in which the narration occurs. The context of utterance however “is set somewhere in the past […], which yields the impression that the (actual) speaker is present at the scene he is describing” (Schlenker 2004: 281). In these cases the present tense is supplied its semantic value from the context of utterance, and not from the context of thought. The context is shifted from its normal context in virtue of the discourse being a narration in the historical present.

There is a very large discussion in the philosophy of language concerning shiftable indexicals and to what extent they do and can occur in natural languages. Recanati himself, for example, suggests that tenses should not be construed as proper indexicals but rather as “perspectivals” (Recanati 2010: 210). I cannot take a stand on all the issues involved here, and it seems to me that not much for my purposes hinges on whether to agree with Schlenker that they are genuine indexicals or with Recanati that they are perspectivals. What I do believe, however, is that the general idea behind shiftable indexicals can also be used to illuminate intentional content.

Schlenker makes a distinction between context of thought and context of utterance. It is possible to make a similar distinction between context of mode and context of (explicit) content. The general idea behind this distinction is that the context of mode is always the context in which a person has an intentional act of some mode. But the context of content need not be identical with the context of mode. In an act of memory, the context of content is shifted from the present to a context located in the past. Similarly, when the psychological mode is anticipation, the context of content is located in the future. In an act of perception, however, the context of content will be identical to the context of mode. The general idea is consequently that context of content is determined by the psychological mode. Whether or not the present tense in explicit content refers to the time of act depends upon which psychological mode entertains the explicit content.

Treating the present tense as it occurs in explicit content as a shiftable indexical has two distinct advantages. First of all, it can explain how memories can be so vivid to us. Schlenker notes that a narrator can use the historical present to great effect in order to yield the impression that the narrator is a contemporary observer of the events he describes. But if this can be said about narrations, it can equally well be said about memories. As I have argued, remembering seeing an event involves re-living seeing the event here and now. But there is also a point of view in the psychological mode with a different context of evaluation, thus explaining how the perceiver is aware that the present of the explicit content is not the present in which the content is entertained.6

The second advantage is obviously that since the context of evaluation of the explicit content is shifted from the present to a time in the past, the act nevertheless has the right kind of conditions of satisfaction. Using shiftable indexicals makes it possible to do justice to the phenomenology of memory, which requires a present tense in the explicit content, and to the conditions of satisfaction of a memory, which requires that the explicit content is evaluated with respect to a past state of affairs.

5 Conclusions

Recanati has in my opinion correctly argued that Searle is wrong to equate content with complete content. If the intuitions underlying the discussion of time consciousness are correct, however, Recanati is mistaken in arguing that explicit content is temporally neutral. This does not entail that the psychological mode makes no contribution to the complete content of the act with respect to tense. If Recanati is correct, a memory of a perceptual experience can in principle have the same content as the original perception. Recanati attempts to locate the difference between the complete contents of the acts in the difference between the psychological modes of memory and perception. This however does not entail that the explicit content is temporally neutral. I have argued that a theory that posits a present tense in explicit content even in memories squares better with the phenomenology involved than Recanati’s theory does. I have also tried to explain how acts of memory can retain a present tense in explicit content and avoid the objections raised against that idea by Recanati.

Footnotes
1

We shall however depart from Meinong in the specifics of the tristinction.

 
2

For a discussion of this point, see Dainton 2000 and Miller 1984.

 
3

As Smith has pointed out, perceptual content seems to be singular and demonstrative rather than indefinite as in Searle’s account (Smith 1986: 201); thus the change from “a red ball” to “the red ball”.

 
4

For the sake of simplicity, this account leaves out the complication that the explicit content also contains a reference to the past tense.

 
5

I am very grateful to an anonymous referee for suggesting that the problem could be treated this way. I owe the entire idea that the tensed indexicals in perceptual content could be treated as shiftable indexicals to him or her.

 
6

I owe this point as well to my anonymous referee.

 

Acknowledgments

Thanks are due to Alexander Almér, Kent Gustavsson, Ingvar Johansson, Christer Svennerlind and an anonymous referee for valuable comments.

Copyright information

© Springer Science+Business Media B.V. 2011