Every once in a while you come across a concept that puts in words something that’s been on the tip of your tongue for years but which you’ve never quite been able to articulate. Ian Hacking’s concept of interactive kinds is, for me, an example of such a concept. In his discussion of social construction talk in The Social Construction of What?, Hacking is careful to emphasize that such talk generally refers to the construction of our categories (kinds), not the individuals or entities that are grouped under these categories. It is not, for example, my cat Tabby that is constructed, but the kind or category “cat”.
What is interesting here is not the construction of categories or kinds themselves, but rather the relationship between the constructed kind and the entities that fall over them. Hacking distinguishes between two different kinds: indifferent kinds and interactive kinds. In the case of indifferent kinds, the entity or individual falling under the kind is indifferent to the categorization that, for lack of a better word, names the entity. Take the example of trees. In his Prolegomena to Linguistics, Hjelmslev teaches us, in good nominalist fashion, to discern how different languages classify trees and shrubs in different ways (cue Monty Python). What is classified as a tree in one language might be classified as a shrub in another language. The important point, however, is that the entities being so classified are indifferent to the classification. The classification makes no difference to them. This doesn’t mean the entities are passive. Plants are active in all sorts of ways. It only means that plants take up no stance with respect to how we classify them.
Read On!
By contrast, interactive kinds are kinds where the entity classified take up a stance or respond to the manner in which they’re classified. There is an interaction between the classification and the entity. A person classified as an alcoholic by a doctor can take up a stance with respect to this classification. They can contest the classification. They can begin to emulate features of the alcoholic (“Hi, I’m Regi and I’m an alcoholic!”) and so on. The classification doesn’t merely describe but begins to influence the behavior of the person classified, thought, and self-experience of the person classified.
Lacanian psychoanalysis is particularly sensitive to how classifications function as interactive kinds. In “Position of the Unconscious” Lacan argues that the unconscious has a history and changes in response to psychoanalytic interpretation. Here he’s not talking about individual analyses, but collectively. Symptoms take different forms at different points in history (eg, we no longer see the dramatic symptoms of hysteria witnessed early in the last century. Hysteria now produces new symptoms). Lacan hypothesized that the unconscious collectively responds to the manner in which it’s theorized such that it finds new ways to repress desire when desire is brought out in the open. Similarly, standard interpretations become stale and no longer produce shifts in symptoms. Oedipal interpretations no longer produce the dissapative effects they once did. Likewise, Lacan elsewhere suggests that analysand’s have dreams in line with the theory of their analyst (as a result of transference). Patients of amLacanian will have linguistically structured dreams, Jungian patients dreams rich in symbolism, Adlerians…, etc. Finally, if Lacanians do not reveal their diagnosis to their patients (“you’re an obsessional”) then this is precisely because diagnostic categories are interactive in this way. The diagnostic category risks becoming as an impediment to analytic progress as it becomes a script for the patient as well as a universal causal explanation (“I do x because I’m an obsessional!”), thereby delaying or foreclosing an encounter with desire.
Nonetheless, it would be a mistake to conclude that the interactivity of interactive kinds is restricted to the stance a person takes up with respect to the category. Classifications are social entities, material realities, embedded in both people and institutions. When a child is categorized as ADHD this activates all sorts of institutional protocols, leading schools and insurance companies to behave differently towards that child, but also family members and other persons in that child’s social world. That child might not have the faintest idea that he’s been classified in this way, but nonetheless the classification has all sorts of effects on the child because of how others relate to him.
The most appealing feature of Hacking’s analysis of interactive kinds is that it doesn’t treat persons as passive bearers of the classifications that descend upon them. It’s a two-way and interactive street. In critical theory we talk a lot about subjectivization, suggesting that persons, subjects, are just effects of classificatory system in which they’re enmeshed. Here we repeat the Aristotlean hylomorphic model, conceptualizing persons as passive and unformed matters that are given form or structure by classifications. However, while we can certainly internalize classificatory schemes, thereby being subjectivized, more often than not classifications or kinds are a bit like a sticky web that poses all sorts of constraints on how we navigate the world without fully forming us. We can push back in all sorts of ways,leading the schemes themselves to change and shift. In Lacanian terms, this would be the difference between the university discourse and the discours of the hysteric. The university discourse dreams of perfect subsumption under the signifier or knowledge, producing a divided subject as a result (the individual becomes enmeshed in categories that structure their life while exceeding their ability to master them). In the discourse of the hysteric, the categories (S1) are contested by the alienated subject. What’s produced? New categorizations. The Butcher’s Wife has a dream where her wish is not satisfied. What’s she saying? “Take that Dr. Smartypants! How can your wish-fulfillment account for a dream where the wish is not fulfilled!” Freud thereby finds that he must create new signifiers to account for this anomaly.
January 27, 2011 at 10:46 pm
I can see OOO being a more expansive treatment of memetics and other idea theories. The meme theory never really took off for precisely the reasons that have plagued contemporary thought: on one side of the divide is the philosophers in love with language and signs, on the other side is the scientists in love with a crude ‘materialism’. Neither is willing to give autonomy, let alone reality, to memes/idea-objects. OOO could adopt some of the techniques and ideas of memetics, as well as the insights of Hacking to make something more durable.
January 28, 2011 at 3:39 am
Wow, this is fantastic. I was just thinking about why interpretations in therapy can be so intense and this nails it.
January 28, 2011 at 3:53 am
Hacking’s little book is good stuff.
January 29, 2011 at 1:17 pm
Hacking is giving us ‘good stuff’ but it is possible (and Hacking is pretty careful about this) to over value the impact of the label(ing) itself and miss the ways in which diagnoses are more dangerous/limiting in that the diagnosis becomes understood, and acted on, (primarily first by clinicians and other professionals)as the sole motivation for, and therefore answer for (including speaking for the person/subject) for, interpretation of, all of a persons doings, and as a ‘scientific’ (evidence-based) justification for all ‘treatments’ and restrictions (such that providers are lead by (just following)science and not their own/owned choices/interpretations to get back to the limits of Harris), we render people one dimensional (they are a schizophrenic and not person who suffers with schizophrenia) and in R.Ellison’s sense invisible, as we pre-define them and so don’t have to attend to their broader/multidimensional humanity. Diagnosis becomes the end not the beginning of the inquiry process making people into manageable units. Wittgenstein and his followers are pretty good at noticing such suggestive/reductive powers of analysis. On the broader point of the powers/potentials of suggestion/lures see Stengers on a Constructivist reading of Process and Reality.
January 29, 2011 at 4:42 pm
dmf,
I discuss exactly this in the post.
January 29, 2011 at 9:33 pm
you do sketch out some aspects of these problems, I was trying to flesh out some of the implications and offer a possible constructivist/pragmatist alternative.
January 29, 2011 at 9:37 pm
dmf,
Hacking is a constructivists. He’s analyzing exactly the sorts of things you’re talking about so as to raise the sorts of questions you’re interested in. The point is to allow critical analysis of how those categories function socially.
January 29, 2011 at 10:14 pm
Hacking is indeed a constructivist, but is Lacan? Not to my understanding and so I offered Stenger, though when it comes to analysis as actual interventions in systems/institutions I prefer Rabinow.
January 29, 2011 at 10:47 pm
Lacan’s not a constructionist? That’s a hard case to make with his analysis of the role of the signifier in subjectivity. Take the example of the two doors, male and female, in “Agency of the Letter”, or his claim that men, women, children, etc., are effects of the signifier in Seminar XX. He’s extremely attentive to the realities that we construct through language. He believes there are real things, certainly (mathemes, structures). So does Hacking. Either way, I take useful concepts wherever I find them.
January 30, 2011 at 12:46 am
dmf,
I should also add that “constructionism” is not a bad word for me. I consider myself a realist constructivist. I believe a number of things are constructed or built that wouldn’t otherwise exist. However, I also see these constructions as material processes that don’t necessarily involve humans at all. Lacan tends in the social constructivist direction. Remember in Television and RSI he treats reality (not to be confused with the real) as an effect of the symbolic and the imaginary. Remember also that the symbolic is the domain of the signifier. The key feature of the signifier is that it is arbitrary. I am more than happy to embrace the reality of the signifier (I’ve argued repeatedly for it here). I cannot, however, endorse the thesis that beings are constructions of signifiers. Rather, I see signifiers as entities in their own right that interact with a variety of other entities.
January 30, 2011 at 2:07 pm
certainly inside, and outside, of the mangle of our practices there are the physical processes of the universe, but we part company on the reality/independence of signifiers and structures of Language, to my experience/understanding even Annmarie Mol’s local ontologies are too stable/unified and I’m more taken by the possibilities of (post)phenomenology/enactivism. We certainly don’t need to rehearse the structuralist/post-structuralist debates here but if you get a chance check out Stengers.
February 6, 2011 at 1:54 pm
[...] something relevant to what that blog will be my attempt to think through in relation to psychiatry. Larval subjects on [...]
February 6, 2011 at 1:57 pm
I’ve been thinking in terms of idea-objects, although I’m yet to write anything about it. I’m beginning to wonder if the unconscious isn’t the name of the submerged aspect for ideas, where they interact with one another. Perhaps I’m afraid of the ‘idealism’ bogey-man.
February 16, 2011 at 2:23 pm
[...] Bryant, Interactive Kinds *Categories and how people negotiate given identities. A different perspective on how to view [...]