LATOUR’S CATEGORY MISTAKES: tendentious language and doxic retranscription

In AN INQUIRY INTO MODES OF EXISTENCE Bruno Latour describes the felicity conditions of  the diverse modes of existence that characterise the Moderns in an attempt to dissolve certain problems and disputes by showing that they are generated by the illegitimate amalgamation of criteria and of expectations constitutive of incommensurable modes of existence. Such amalgamation gives rise to « category mistakes » where beings or utterances of one mode are interpreted and made to act in terms of the grammar of a quite different mode, thus producing paradox, nonsense, and paralysis.

Unfortunately Latour does not seem to be aware that certain words he uses as part of his theoretical metalanguage potentially ensnare us in problems that could have been avoided by employing a more careful vocabulary. Such is the case with the word « access », which strictly speaking should be reserved for the mode of reproduction or of reference or their crossing, but should not be employed in the philosophical metalanguage.

Philosophy seems to pose a problem for Latour, and is not officially included in his list of modes of existence, but is declared to be definitory of the Moderns: « the Moderns are the people of Ideas; their dialect is philosophy » (22). Beyond this existence as omnipresent within the object of study, the Moderns, philosophy is also an essential part of the theoretical apparatus that describes that object: « I have not been able to see any way this inquiry into modes of existence could do without philosophy. I am turning to philosophy, then … in the hope of forging a metalanguage that will allow us finally to do justice, in theory, to the astounding inventions that the fields reveal at every step—among the Moderns too » (21). Thus even the term « philosophy » suffers from an ambiguity as to the level of analysis it invokes (object-language or metalanguage) and so is potentially misleading. Latour in fact needs philosophy to be a « metalanguage » rather than a mode, because the whole thrust of his argument is to prevent any one mode as setting itself up in totalising hegemony over all the other modes, imposing its own terms and categories as obligatory metalanguage for all the other modes: « no mode can be singled out to contain all the others, to serve as a metalanguage for all modes » (316). Philosophy as metalanguage is not a mode.

For similar reasons, it would have been better if Latour had avoided the vocabulary of « access ». Fortunately Latour is sparing in its use (as compared to Harman) but he uses it far too often, to talk about objective knowledge and the mode of reference, and also about the relation to the neighbor and the religious mode. This is a categoreal error: knowledge is not access, that idea is a version of the naive empiricist vision that denies the ineliminable collaboration of mediators in the knowledge process. « Access » is a bad metaphor for the knowledge relation, but it is an even worse metaphor for relations of human proximity. Reducing caring for the other to a matter of « access » constitutes an illegitimate translation of religious categories into those of reference.

A related pair of terms that illegitimately straddles incommensurable modes is that of « near » and « far », or « close » and « remote ». In a striking inversion Latour claims that religion is not separate from life, occupied with belief in a distant supernatural sphere. This trait, that of dealing with far off, remote objects, is attributed to knowledge (whose legitimate mode is that of reference):

« The work of reference … relies on the establishment of a series of transformations that ensure the discovery and the maintenance of constants: continuity of access depends on discontinuities. This is the only means … to ensure the back-and-forth movement, the coming and going owing to which one can start from a given point (a laboratory, an institute, a computer center) and reach another, more or less remote. Think about the hundreds of successive operations required by an electron microscope through which a researcher ensures access to the division of a cell that cannot be seen by the naked eye. Think about the strings of calculations needed for the spectrum analysis owing to which an astronomer ensures access to a galaxy, also invisible to the naked eye. Two infinities that should not scare us, since biologists and astronomers both have access to them without the slightest vertigo, from their laboratories » (107).

Obviously a distant galaxy is « remote », but that has no bearing on whether science occupies itself in its essence (as defined by its constitutive felicity conditions) with the remote. The atoms composing my body are not “remote” in any non-question-begging sense. Nor is gravitation, nor the neuronal unconscious. « Remote » as a term belonging to the mode of reference is not at all the same as « remote » employed, infelicitously in my opinion, in the philosophical metalanguage. In the example developped at some length in Chapter Three, the hike to and up Mont Aiguille, the chains of reference do not distance us from the mountain summit but lead us to the point where it is under our feet and all around us. Visibility to the « naked eye » is an empiricist criterion, and has no bearing on metalinguistic definitions of proximity and remoteness. (NB: it should be noted that Latour glibly makes reference to « two infinities » to reinforce this sense of remoteness, when in fact no infinity is referred to).

Latour consigns science to the realm of the remote as part of a strategy of special pleading for the religious mode, presented as occupied with that which is closest: « see what is close at hand, our neighbor, the present, the here and now of presence, the only promises that religious words can actually keep: that of incarnation in time, provided that we never exit from time, that we start over and over » (320).

Latour’s grammar of religious utterance is incoherent through trying to include too much. For example the emphasis on « conversion » and the emphasis on « neighbours » are in the same sort of bipolar tension that we have discerned for access, philosophy, distance etc. Latour uses « conversion » to mean turning towards what is closest, the proximate, the neighbouring. Yet it is not obvious, that the « neighboring beings » are « neighbors » in the Christian sense, unless we have already imported that presupposition into our supposedly agnostic and descriptive metalanguage. Even in religious terms it is not clear whether God, as the background of everyday existence, is the closest (cf REJOICING: « the word ‘God’, which once served as the premise of all arguing, could have been translated, when ways of life changed, as ‘indisputable framework of ordinary existence »), or whether one element that is salient in terms of that background, i.e. the neighbour, is closest.

The problem with the concept of “conversion” is that Latour once again oscillates between a common sense acception and a technical, modal, one. Strictly speaking, turning towards what is “nearby” is not something restricted to religious utterance, it is something that both philosophy and science allow us to do. Identifying the nearby with the “neighbor” is tantamount to the category mistake of importing connotations from the doxic domain of religion into the definition of a mode of existence. No reason is given to accord religious utterance primacy in the attention to the nearby, it is presented as « obvious » by Latour. However, this obviousness is an artefact of the methodological and categoreal error that Latour commits rather blithely, considering that his book is in large part a manual on how to avoid such errors.

Finally, no argument is given for accepting the « religious » as a mode having its own singularity and unity. There is no reason to believe that such diverse specifications as attending to God, attending to angels, attending to neighbours, producing unified persons, bringing persons from death into life, can be lumped together in a single homogeneous grammar constitutive of a separate unitary mode.

Cet article a été publié dans Uncategorized. Ajoutez ce permalien à vos favoris.

Un commentaire pour LATOUR’S CATEGORY MISTAKES: tendentious language and doxic retranscription

  1. dmfant dit :

    « There is no reason to believe that such diverse specifications as attending to God, attending to angels, attending to neighbours, producing unified persons, bringing persons from death into life, can be lumped together in a single homogeneous grammar constitutive of a separate unitary mode. »
    I would say that there is no necessity to any of the attempts (some quite recent) that people make to write differing activities into/under « modes », they may or may not serve certain pragmatic interests as prototypes/perspicuous-re-mind-ers/tools/etc but we should judge them by their efficacy in relation to those goals/interests.
    for an interesting and therefore « strong » mis-reading of Whitehead along these lines see:


Votre commentaire

Entrez vos coordonnées ci-dessous ou cliquez sur une icône pour vous connecter:


Vous commentez à l’aide de votre compte Déconnexion /  Changer )

Photo Google

Vous commentez à l’aide de votre compte Google. Déconnexion /  Changer )

Image Twitter

Vous commentez à l’aide de votre compte Twitter. Déconnexion /  Changer )

Photo Facebook

Vous commentez à l’aide de votre compte Facebook. Déconnexion /  Changer )

Connexion à %s