language games, naiveté and organization theory
Teppo
For several years now I have been having a running debate and discussion (and friendship) with JC Spender about various issues related to organizations. We rarely agree, but always have fun. And, invariably we end up in various metatheoretical and epistemological debates, usually with JC pulling out the ultimate trump card: “language games.”
Here’s a (slightly edited) response that I recently wrote to him (perhaps more on the context later):
Now, I am of course not completely naive. I recognize that theoretical insights from different disciplines can be, and often are, contradictory, sometime wildly so, as illustrated by the re-emergence of the neoclassical economics versus organization theory clash (Ferraro et al. 2005; cf. Pfeffer, 1997). But, being the naive realist that I am, I believe that these clashes can and ought to be settled via the merits of the respective arguments rather than merely referring to ‘language games.’ In other words, we must realize that reasoning and science relies on that “massive core of human thinking which has no history” (Strawson, 1959: 10). Thus, referencing language games is simply an act of academic cowardice — an effort to avoid truly engaging with the issues, logic and arguments at stake. Furthermore, citing language games inherently does not recognize that some theories and arguments and facts simply are false, period. And, vetting the ‘true’ and ‘false’ of arguments, proposed facts and theories is at the very heart of science.
Pulling the card of “language games” is a special case of a general argument for the “complete incommesurability of worldviews.” This argument has various precedents in continental social theory, anthropology and political theory (where as Dick Pels argues, was actually refined by the radical right; i.e. Carl Schmitt, but is most clearly recognizable to us in its “cultural relativist” left-wing anthropological form and in feminist and postcolonial theory as the “standpoint epistemology” argument). In sociology, the locus classicus is Durkheim and Mauss on Primitive Classification, in analytic philosophy it begins in the philosophy of language (the phrase of course comes from the late Wittgenstein) and Quine on the “impossibility of translation.” This was clearly the reason why Kuhn’s Structure was such a sensational (and in a Medici like sense “multivocal”) book, since when it came all of the elements of the zeitgeist were pointing in that direction.
It is clear to most people that have seriously thought about this, that there is something deeply wrong here. The problem is that this argument cannot be resolved at the metatheoretical level. At that level the “language game/incompatible worldviews/social construction of collective realities/socialized cognition” will always seem to win. Skeptics are reduced to inarticulate stompings of feet of the “I refute it thus” variety, but I have not seen any argument at a theoretical (not epistemological, or metatheoretical) that pulls the rug back in a adequate manner. Hence we live in the uneasy equilibrium that in our practical experience we all realists of some kind or another (that is we don’t really believe that experiences are radically incommesurable), but in our theoretical self-reflection we all write the front-end of our papers by producing some weak (sometimes strong) version of the language games argument.
Omar
February 14, 2008 at 11:20 pm
I guess I don’t understand the premise of the language game card. What is it about the language of a particular perspective that makes it inimical to translation? Isn’t that the purpose of being an educator – to figure out how to translate your language to diverse audiences and therefore make it more commensurable?
brayden
February 14, 2008 at 11:27 pm
But Wittgenstein wasn’t using the blended space metaphor of language games to refer to either natural language or games. He was in a sense rediscovering the notion of “background” that had played such a key role in continental philosophy from Husserl to Heidegger (and moving to sociology) Schutz and Berger and Luckmann and which survives in institutional theory in the idea of the “taken for granted.” In this way language games refers to implicit know-how, not explicitly articulable rules, so they cannot be translated into one another, since there is no dictionary with which to match “terms” one to one.
Omar
February 14, 2008 at 11:32 pm
“Language games will always win.”
Where are these games winning?
Perhaps the fact that there is no convergence of disciplines can be seen as a victory of sorts for heterogeneous linguistic spaces and language games, but, I think each game and sub-game also is addressing a specific non-overlapping niche or phenomenon. To the extent to which there are competing claims about similar spaces and phenomenon (recent interesting debates e.g., at the nexus of biology and sociology), thats where the rubber hits the road, right? Over time, someone’s claims (because of the claims and not because of their socio-political-linguistic machinery — or?) are better. We now have hundreds of years of history to point to the many junk claims and sciences, long discarded given that better, truer etc explanation emerged. So, perhaps language games exist in the short run, but in the long run we get better and better explanations and theories.
Hopelessly naive, I know.
tf
February 14, 2008 at 11:52 pm
I do want to point though that people pull the language game card without ever recognizing it. At least JC is being explicit about his belief that some languages are incommensurable.
A common experience I have had when talking to organizational economists is that I make an argument, they offer their interpretation of what I just said, I explain how it’s not the same, and then they explain how their interpretation is better because it’s based on X assumption. Okay, but how do we reach convergence if I’m not willing to assume X? Perhaps this is the kind of language game that Omar is talking about?
brayden
February 15, 2008 at 12:03 am
Teppo, do all your personal emails have citations?
Fabio Rojas
February 15, 2008 at 12:33 am
Hah — yes, its how I talk to my wife now too.
tf
February 15, 2008 at 12:35 am
And, I don’t use question marks anymore either…like Quine (or was it someone else?), everything I say is a statement of fact!
tf
February 15, 2008 at 12:37 am
There are no relativists at 30,000 feet – Richard Dawkins
stevphel
February 15, 2008 at 10:22 am
brayden Says:
“A common experience I have had when talking to organizational economists is that I make an argument, they offer their interpretation of what I just said, I explain how it’s not the same, and then they explain how their interpretation is better because it’s based on X assumption. Okay, but how do we reach convergence if I’m not willing to assume X? Perhaps this is the kind of language game that Omar is talking about?”
Classical physics. An assumption: mass is constant.
Relativistic physics. An assumption: mass changes with velocity.
Someone who belongs to the bowling league throws a bowling ball down the alley. He insists that the velocity of the ball has no effect on its mass and he can demonstrate that in the alley.
The physicist can demonstrate that an atomic particle accelerated close to the speed of light will gain a great deal of mass. He states that the bowling ball gains a very,very, tiny, tiny amount of mass.
The bowling league association is just wrong. But it has no practical effect on bowling.
The Quantum Physicist says they are both wrong.
The x-th tuple Physicist says groups 1 through x-1 are wrong.
They are all wrong depending on the context. If any of them can be demonstrated to be practical in a particular context they will endure. If the bowling league organization approaches the speed of light, they will collapse on themselves and disappear in a puff of smoke.
Doug
February 15, 2008 at 11:13 am
“…being the naive realist that I am, I believe that these clashes can and ought to be settled via the merits of the respective arguments rather than merely referring to ‘language games.’ In other words, we must realize that reasoning and science relies on that ‘massive core of human thinking which has no history’ (Strawson, 1959: 10).”
I like this way of putting it. I think language games worked best for Wittgenstein when he used them as thought experiments not social ontologies. But even Wittgenstein would say that since language games have no explanatory power (we can only use them to describe language) they all have no normative force. So the appeal to language games can’t end the discussion (as tf rightly suggests when he asks, “Where would they win?”)
That said, I think cowardice is sometimes justified. Or rather, we might compare the “cowardice” of invoking language games with the “bullying” of invoking “the massive core of human thinking which has no history”. Neither really engages with whatever one might be discussing. Both are ways of avoiding actually having to win the argument.
A gesture at the ahistorical core of your argument is not “vetting the true and false”. Perhaps Wittgenstein is right after all: the core is only the core if it remains unspoken. Invoking it is just another move in a language game.
Thomas
February 15, 2008 at 1:25 pm
[...] Last week I raised the issue of “trump cards” in scholarly dialogue and debate, specifically the matter of “language games.” [...]
another trump card: the nobel prize « orgtheory.net
February 26, 2008 at 12:50 am
brayden: “Okay, but how do we reach convergence if I’m not willing to assume X? Perhaps this is the kind of language game that Omar is talking about?”
Presumably we move down onward until we arrive at some assumption we can agree upon, and build from there? This has always seemed the most obvious resolution to me.
Naturally, it’s neither perfect, and the talk of assumption’s not meant to suffocate it within a purely deductive framework. The first, relevant common assumption in itself might be (and with particularly distant worldviews probably will be) something only infinitesimally useful in itself, and require a chaining together with many other ‘common assumptions’ to patchwork something together which could potentially cause two rational actors’ views to converge, but presuming that two worldviews are *completely* incommensurable seems quite absurd; it’s not at all difficult to think up endless examples of universal human assumptions, though it’s probably often harder to state them very rigorously.
An incommensurability argument with more meat to it (imo, naturally) would focus at the intermediate levels in the above chain of assumptions-deductions-inductions and argue that it is simply highly implausible (say, with probability P in X years) that two rational actors with sufficiently distant worldviews will ever come to recognize their common assumptions. Could make for some entertainingly wild sample space constructions…
Coming up with natural units for “the distance between two worldviews,” or establishing bounds on the diversity of inductions-deductions-assumptions, half accounting for less prescriptive, more realistic ideas about human thinking seem well beyond our ken now, if ever — but conjecturing that we consider the incommensurability of two worldviews largely by analogy to probability theory, dynamical systems theory, etc., ignoring the difficult details, succeeds in making the point that the difference between two worldviews need not necessarily be binary or nominal, and that if it’s not, the implications are.. well, not astoundingly concrete, but pleasantly open. A slightly different worldview for considering worldviews?
praxeologicalGoertzelite
March 8, 2008 at 3:20 pm