Spectators of the ongoing U.S. political circus will have noticed that Senator John McCain has started referring to his opponent, Senator Barack Obama, as “That one.” This got me thinking about theories of reference and their function in everyday discourse, particularly as a form of address. Recall how Bertrand Russell distinguished what he called “ordinary” proper names and “logically” proper names. “This” and “that” are examples (perhaps uniquely) of logically proper names. By contrast, ordinary proper names are abbreviated definite descriptions. “Aristotle”, consequently is simply shorthand for “The teacher of Alexander the Great” or “The author of the Nicomachean Ethics” or some conjuntion of two or more such descriptions (descriptions, that is, of the form “The…”) McCain’s reference to Obama as “That one” replaces his previous habit of referring to his opponent as “The one.” A transition, in other words, from a definite description to an indexical; from an ordinary proper name to a logically proper name. Could the political motivations for this switch of rhetoric be explained by philosophical theories of reference?
The original form – “The one” – was, no doubt, intended to encourage public perception of Obama’s messianic pretensions; of someone who symbolises – not merely eulogizes – what the Americans call ‘hope’. The ideas of hope, change, unity themselves, rather than just the personal character and integrity of a single man. McCain, by contrast, has presented himself as a man of extraordinarily unique character: a war hero, survivor of a P.O.W. camp, an experienced politician, and, crucially, as a “maverick”. Whilst Obama’s U.S.P. has been his universality, his celebrity, McCain distinguishes himself as unique, singular. “That one”, usually accompanied by the obligatory finger-point or dismissive gesture, is simultaneously disdainful and grounding. It identifies its referent specifically and rejects any symbolic description supervenient upon that identification.
This, only slightly tongue-in-cheek, gambit is my way of introducing a socio-political theory of reference, whereby individual bearer’s (e.g. of a name) acquire that name by virtue of their status. The problem of reference is epistemological to the extent that it questions whether our referring terms correspond to specific external entities and processes, and, if so, how this can be done and how it can be known to be done. Much of our knowledge implicitly refers to bearers who may be tracked through space-time without the need for a new referring term. Thus, I am the same individual I was five years ago, am today, and will be in five years time. We cannot understand the use, development and evaluation of knowledge without a workable solution to this problem.
My thoughts on this matter have been strongly influenced by having just finished Barry Barnes‘ (a former faculty member here at Edinburgh) Understanding Agency (2000. London: Sage). The book is a compelling entreatment for a move away from individualistic theories of free will and agency towards an empirical, naturalistic, and thoroughgoing sociological study of human beings as “highly gregarious, interdependent social primates.” (Gagnier and Dupre, 1998. “Reply to Amariglio and Ruccio, in M. Osteen and M. Woodmansee (eds.) New Economic Criticism. New York: Routledge.) A sentiment expressed, believe it or not almost verbatim, by my late maternal grandmother. It should be required reading for any student of moral philosophy, being concerned with the previously intractable dualism of causal and voluntaristic explanations of agency, but its epistemological implications – Barnes being one of the key movers in the sociology of scientific knowledge – is of equal utility. Although the book is written within the social theory tradition its debt to a certain empiricist philosophical tradition is evident (Hume and Locke loom large). Further, and of most interest here, the final chapter contains an injunction for the value of Kripke’s Naming and Necessity (1980. Cambridge: Harvard University Press) to sociology.
Recall Saul Kripke’s claim that ordinary proper names are rigid designators. Since it is not a necessary property of, say Aristotle, that he did author the Nicomachean Ethics, or any single one of the many descriptions that are ascribed to him, then it cannot be that a set of descriptions uniquely ‘picks out’ the referent ‘Aristotle’. Aristotle is ‘Aristotle’ across all possible worlds. Now, personally, I, with Quine, am skeptical of Kripke’s possible world semantics but can, despite this, acknowledge the great insight of Kripke’s theory. For Kripke, it is not the empirical characteristics or describable properties of a human being that make her an individual of a specific name. Human bodies do serve as markers through which ‘the same individual’ may be monitered from instant to instant by a community. However, what constitutes ‘the same individual’ over time is not her empirical characteristics. No physical changes to an individual, no matter how radical, nor changes in behavioural or dispositional chracteristics, dispossess an individual of her lifelong association with her name, or otherwise count her another person. How, then, do we attribute names and recognise their bearers? How do we know who she is?
Kripke defines an individual not by her nature or any constancy therein, but as a continuous line through space-time. His causal-historical account of naming ‘baptisms’ describes the ascription of a name to whatever persists in spatio-temporal continuity with the object of that baptism, in much the same way that a name-label is tied around a bird in an aviary. The naming of an individual is a virtual name-label, constituted of ongoing speech-acts, and may be used by the community similarly to refer to that individual. Allow me to quote Barnes in full:
In this way naming activity, and indeed the entire institution of the naming of individual persons, may be understood as a collective accomplishment through which a physical body and an associated essence are tracked over a life-course, and kept visible as entities we orient ourselves toward. Addressing this account naturalistically, it implies that what actually allows identification of ‘the same’ human being in a changing body is not the essence of an individual within but continuing references to that essence from without. It is the ring of speech and action around the body, no particular part or component of the ring but merely the persistence of such a ring over time, that constitutes the human being qua individual. If the ring disappears, then the body becomes a mere material object and its status as an individual lapses. (2000, p. 147)
Read alongside Barnes’ “Social Life as Bootstrapped Induction” (1983. Sociology vol. 17, no. 4, 524-545), the problem of reference for sociology of scientific knowledge is brought out. ‘Bootstrapping’ is a metaphor that we are very familiar with in philosophy and also, I am informed, has applications in computing, genetics, and is used in a Robert A. Heinlein novella about recursive time travel! Barnes’ concept of bootstrapping also retains this reflexive implication. For him, acts of reference and acts of inference are connected by their reflexivity: much referring is self-referring, and much inference is self-validating. This takes place insofar as inductive inferences become “permeated with feedback-loops or ‘bootstraps.'” I first became pre-occupied (appropriately enough!) with recursivity through Douglas R. Hofstadter’s Gödel, Escher, Bach ( 1999 . Basic Books). The aleatory connections between Barnes’ work on bootstrapping and Hofstadter’s book suggests reading the two together. Barnes considers the possibility (or even desirability) of removing/avoiding these feedback-loops or reflexive inductions. In some cases, for example the self-fulfilling prophesy, induction is destructive. In others, bootstrapped induction plays a constructive role in constituting stable institutional forms.
One might assume that the problem of reference would be easy for the natural sciences to solve, being concerned entirely with “inert, objectively accessible physical phenomena”. (Barnes, 1983) But even here there is no widely accepted theory of reference. And one can be sure that if the problem is difficult for the natural sciences, it is much more problematic for everyday speech which so often is about speech itself. Thus, speech acts are referring acts which are at the same time referring – reflexively – to other speech acts. This being the case, it is impossible, unlike in scientific discourse, to clearly delineate references from referents. Hence, it is not possible to specify what the community has knowledge of independently of what the community knows of it. The discourse of the community, in other words, is partially self-referential and its knowledge includes a self-referential component.
Barnes makes a distinction between two extreme potential types of referring terms: N-type terms (for ‘natural kind’) and S-type terms (for ‘social kind’). N-type terms, such as ‘tree’ or ‘leaf’ are acquired through perceptual encounters with the empirical characteristics of a leaf. An individual approaches an object they take to be a tree. As she moves closer she expects, by routine induction, to encounter what she takes to be leaves. The usual stereotype for the use of, say, ‘leaf’ is as a kind of pattern recognition. The agents nature and socialisation affect how he acquires a pattern for ‘leaf’ (what she has been shown of leaves, told of them, etc.) It is then the qualities of the putative leaf (its oval shape, its green colour, its waxy texture) that are usually sufficient to decide whether or not it matches the pattern.
A system of verbal knowledge constituted entirely out of idealised N-type terms would contain no self-referential features (this is, perhaps, how we tend to view extant scientific knowledge). The other extreme of this account Barnes calls S-type. ‘S’, in contradistinction to ‘N’, is not used on the basis of empirical features of its referents, and so it must be used on the characteristics of the users of the term: its application must be entirely a matter for their judgement, their decision, their agreement (appropriateness?). There are plausible candidates for this kind of term: consider a situation where you walk into a lecture theatre and see an individual standing behind a lecturn and other individuals sitting in rows before her. This community’s use of ‘lecturer’, ‘student’, etc. has as little to do with empirical characteristics as the identification of a man as ‘married’ has anything to do with his wearing a ring. In practice, it is no doubt a bit more complicated that this, but one might follow a crude interpretation of J.L. Austin’s performative utterance whereby the declaration ‘this particular is an S’ instantiates that particular an ‘S’ by fiat.
Bootstrapped induction is a process whereby individuals where individuals make inductive inferences about entities, which collectively, by their inferences, they create and maintain. If anyone is interested in a fuller (though quite technical) account of the ‘designation devices’ that illustrate this reflexive induction, I would encourage them to read the aforementioned article. Barnes even claims that there are epistemological systems which hold that inductive procedures can be dispensed with altogether. I am not sure which authors he is referring to there and he cites no sources in support of this claim. Can anyone help me out?
[The Barnes-related half of this article is based on a presentation I gave to the Sociology of Scientific Knowledge seminar at the University of Edinburgh Institute for the Study of Science, Technology and Innovation, 09/10/08.]