Time, Homogeneity and Phenomenology
Time, Homogeneity and Phenomenology


The relevance of Wittgenstein's reflections on time (to be found in chapter seven of the Philosophical Remarks) for the understanding of Wittgenstein's dismissal of the project of a phenomenological language has already be noticed by commentators like Denis Perrin (Perrin 2004) and Mathieu Marion (Marion 1998). Nevertheless, I think they do not provide the complete understanding of the reasons that lead Wittgenstein to this dismissal. In this paper, I will sketch a brief analysis of this chapter, in which I try to show that a central clue to this issue is found in Wittgenstein's mention of the Bergsonian conception of the time of immediate experience as 'heterogeneous', in opposition to the 'homogeneous' physical time.

Table of contents

    For brevity's sake, I will take for granted some points which certainly should deserve a more detailed discussion. These points can be summarized by the opening sentences of the Philosophical Remarks. There, Wittgenstein says that now he does not have phenomenological language as a goal, just after having said that a proposition is completely logically analyzed if its grammar is made completely clear, no matter what idiom it may be written or expressed in. It seems to follow from these two sentences taken together that the construction of a phenomenological language is tantamount to a complete logical analysis of our language. What Wittgenstein is abandoning then is the idea that this logical analysis must result in the construction of a special language (a language whose notation obeys to the logical grammar - as in Tractatus 3.325 -, in which to each object of thought there corresponds an element of the propositional sign - as in Tractatus 3.2). If this is correct, then the opposite of a phenomenological language, the hypothetical or physical language or, more simply, our language must be conceived as a language whose propositional signs do not specify visibly, so to speak, the form of their sense. Now let us turn to what seems to be the rationale for this rejection: the analysis of time ('primary time', as Wittgenstein says, but also 'memory's time') reveals that it is impossible to 'picture' the temporal determinations. This result is to be found in the seventh chapter of the Remarks.

    We can distinguish in this chapter VII three different sections. The first one (from paragraph 67 to 69), shows - on the basis of the description of what could be termed 'the best phenomenological language possible' - that it is not possible to eliminate completely the hypothetical aspect - a language such this just described would still be 'hypothetical' and any thing 'below' it would just cease to be a language. The final conclusion of this first section shows clearly that it is the temporal aspect of the world that renders impossible the construction of a phenomenological language: 'Isn't it like this: the phenomen (...) contains time, but isn't in time (...) Whereas language unwinds in time. // What we understand by the word 'language' unwinds in physical, homogeneous, time' (Wittgenstein 1975, p. 98; Rhees omitted the term 'homogeneous' - cf. Wittgenstein 1994, and the page 28 of Typescript 209 , in Wittgenstein 1998 - but there is no reason to do so, and this word is important, as we shall see).

    Then follows a second section (from paragraph 70 to 74), of a more positive character, introduced by an 'on the other hand...'. On the other hand, I can realize the analysis with respect to the other aspects, although not the temporal one. Thus, starting from the 'physical' representation of the Euclidean space that extends out around me and in which I can locate myself, analysis will show that it consists in a 'mixture of visual space and the space of muscular sensation'. As a result of this analysis of my perception of space, I discover that in it we cannot find an 'I' to which it relates, that this 'I who perceive' is the confusion between two different spaces (visual and muscular). The origins of this analysis are known: they go back to Berkeley, but they are renewed in the time of Wittgenstein by authors as Mach and Poincaré. But, more interestingly, it is impossible not to recognize in this section of chapter VII something as a version in extenso of the aphorism 5.631 of the Tractatus: it starts describing 'the world as I found it' and concludes by showing that in this description there's no mention of an 'I'. Its purpose, then, seems to be to indicate - quickly - that the project of a complete analysis is disturbed only by the temporal aspect, and that it can be pursued with the other aspects of reality.

    Aphorism 75 opens the third and last section of chapter seven. The analysis of visual space was completed, and now Wittgenstein seems to come back to the subject of the possibility of a phenomenological language (although this possibility was already set aside). This section starts asking how phenomenological language treats time, and goes on considering a new 'model' of a phenomenological language.

    Let us review quickly the first section. In it, I want to stress three points. Wittgenstein starts assuming that he could remember all its sense-data. In that case, he asks, what would prevent me from describing them in a non-hypothetical way? My first point, then, is this assumption of an absolute memory.

    I could describe the visual images that I saw in the following way: through plaster-cast figures on a reduced scale and shaped only until the point where my vision had effectively reached (the back of these little statues, for example, would not be shaped, and they would be shaded as inessential). The second point, then, is this 'symbolic negligence': although this description tends to a natural syntax (the round symbolizing the round, the red symbolizing the red, etc), there is no need to be strict here.

    Until here, says Wittgenstein, everything would be fine. But then [he] asks: what about the time I take to make these little statues? He assumes that they would be ready in the same speed in which my memory goes. This is the third point, that contrasts with the second. There - in the case of the visual aspects -, we had no reason to be strict - the reproduction could be made in reduced scale, with some parts blurred by a shading as inessential; here, with regard to time, he excludes the idea of a scale: what he calls for is the same speed.

    Finally, Wittgenstein concludes: but let us assume that I read again this description - isn't it now hypothetical, after all? The text is not quite clear, but there seems to be no doubt about the fact that the problem has to do with the temporal aspect; this aspect is emphasized in the third paragraph, just after having said that 'until now everything goes fine' (and, as we said, the end of this first section is quite explicit about the temporal nature of the problem).

    The first question, then, is the following one: why Wittgenstein assumes such a colossal memory? We must first notice that what is assumed is not a colossal trustworthiness of memory. For in the world of the data memory is the source of the concept of 'past', memory gives me immediate access to the past, it is trustworthy ex definitio (cf. Wittgenstein 1975, p. 62 and p.82). What is colossal is the extension of memory. But it will be nonsense here to imagine that this total extension is needed in order to give an integrally truthful description. The phenomenological language is not characterized by only enunciating true propositions, but by specifying its sense completely.

    It seems to me that the only way in which this assumption could be construed is to tie it to the determination of the temporal localization - and, here, the presence of the bergsonian vocabulary of the 'homogeneous physical time' is decisive. If the time of the world of the data - in contrast to the physical time - is heterogeneous, I cannot apply to it a fixed ruler, I cannot determine a certain instant of this time by the repeated application of a certain 'temporal ruler' (something as 'five minutes ago'). The idea, then, which commands this assumption is to determine the temporal position of an event by the insertion of the event-picture that will describe it in a series of event-pictures similar to that one in which the event described is inserted.

    It is not difficult to see what will be the problem, then: such a series of picture-events will not fully determine the flow of experience that it nevertheless tries to describe. What it will fail to determine is, so to speak, the 'rhythm' of this series of events. This failure of the description was not noticed at first because at the moment of its production, it was produced as a copy. At the moment in which I produce one determined statue, when I am 'writing this description', I use as model one definite event that is temporally situated. When I produce the description, I know that this statue represents that past visual image because I am instituting it as a representative of that one. But when, instead of producing the description on the basis of the model, I try to 'read' in the very description the indication of the instant of the past time that is intended by such and such a statue, I do not find this indication.

    But was not the identity of 'rhythm' guaranteed by the fact that the production of statues goes on in the same speed in which my memory goes? What must be noticed is the fact that Wittgenstein here alludes to the speed of my memory and not of the remembered facts, in contrast to what occurred with the visual aspects. There, Wittgenstein does not allude to what I remember to have seen, but to what 'really was seen'; but here the speed is not the speed 'in which they really had been seen', but the speed 'in which my memory goes'. The point is not that perhaps my memory deceives me, but that the memory that 'has a present speed' is a temporal process successive and not simultaneous to the recollected temporal process. To say that it preserves the same 'rhythm' is, strictly, to assume that I can superpose the present time to the past time – that is, to assume that time is homogeneous. So it is the heterogeneity of the time of the data that prevents the figurative language from completely determining its sense. And so, it is this heterogeneity that blocks the phenomenological language: every language 'unwinds in the homogeneous physical time' in the sense that there is a rhythm - a 'durée' - which the language could not fix. This interpretation seems to be confirmed also by the third section of this chapter VII, to which we will now turn.

    As we saw, after having concluded, in the first section of the chapter, that language is necessarily 'something physical', after having showed, in the second section, that except for the temporal determination all the others determinations can be completely pictured, in the third section, Wittgenstein seems to come back to the idea of a phenomenological language (to be true, this move must be differently construed, but such a discussion would take too long ) and to consider a new model of phenomenological description . Wittgenstein starts by saying that we are able to recognize two time intervals as equal. He will assume - to simplify matters - that changes in the visual space are discontinuous and in time with the beats of a metronome. Then, he says, I can give a description of this process.

    It seems clear to me that the equality of time intervals that we are able to recognize as equal refers to simultaneous and not successive intervals of time. That is, what I am able to recognize as equal is the time interval between two successive changes of my visual space and the time interval between two successive beats of the metronome. If this is so, what Wittgenstein is trying with this new model is clearly the change from an absolute determination of the temporal localization to a relative determination of this localization. I specify the temporal position of an event by specifying at 'how many metronome beats' it is to be found of the present moment.

    On the other hand, clearly the absolute determination did not disappear of sight. What symbolizes, here, is the beat of the metronome, and what the relative determination (the 'rhythm') fails to provide per se is introduced, in the description, by the rhythm of the symbol, the rhythm of the beat of the metronome. It is not the simultaneity between beat and visual change that is described: this simultaneity is part of the method of description. Or, in a very simple way: since reproduction of the temporal process was impossible for the fact that two lengths of successive times could not be superposed, we turn to a simultaneous reproduction.

    I hope to have made my point: the bergsonian opposition between the physical, homogeneous representation of time and the heterogeneous time of immediate experience is a central clue to a correct reading of this seventh chapter of the Remarks, and therefore to a correct reading of Wittgenstein’s dismissal of the phenomenological language. But I will end with a few words about the result of the analysis of this new model, just not to let Wittgenstein's argument unfinished.

    Wittgenstein says that in this description everything hypothetical is avoided apart from what is contained in the presupposition that the description is given to me independently of the part of it that is before me now. At first sight, this description determines, by its very temporal rhythm, the temporal rhythm of the process it describes - and if it was so we will have after all some kind of phenomenological language. But this is an illusion: the temporal determination was not pictured. The temporal relations between the various phases of the symbol do not symbolize the temporal relations between the various phases of the symbolized simply because a past phase of the symbol, being past, is not, and so cannot symbolize (as a 'negative fact', in the Tractatus could not per se symbolize - cf. 5.5151).


    1. Wittgenstein, Ludwig 1975 Philosophical Remarks, Oxford: Basil Blackwell.
    2. Wittgenstein, Ludwig 1922 Tractatus Logico-Philosophicus, London: Routledge & Kegan Paul.
    3. Wittgenstein, Ludwig 1994 Wiener Ausgabe, i: Philosophische Bemerkungen, Vienna: Springer.
    4. Wittgenstein, Ludwig 1994a Wiener Ausgabe, ii: Philosophische Betrachtungen, Vienna: Springer.
    5. Wittgenstein, Ludwig 1998 Wittgenstein's Nachlass - The Bergen Electronic Edition, Oxford: Oxford University Press.
    6. Marion, Mathieu 1998 Wittgenstein, finitism, and the foundations of mathematics, Oxford: Clarendon Press.
    7. Perrin, Denis 2004 "Wittgenstein e Husserl leitores de James: a questão do tempo", Dois Pontos vol. 1 n. 1, 59/98.
    Bento Prado Neto. Date: XML TEI markup by WAB (Rune J. Falch, Heinz W. Krüger, Alois Pichler, Deirdre C.P. Smith) 2011-13. Last change 18.12.2013.
    This page is made available under the Creative Commons General Public License "Attribution, Non-Commercial, Share-Alike", version 3.0 (CCPL BY-NC-SA)


    • There are currently no refbacks.