Tuesday, August 7, 2012

Response to Little by Dave Elder-Vass

[Dave Elder-Vass accepted my invitation to write a response to my discussion of his recent book, The Reality of Social Construction (link). Elder-Vass is senior lecturer in sociology at Loughborough University and author as well of The Causal Power of Social Structures: Emergence, Structure and Agency, discussed here.  Thanks, Dave!]

Social construction and the reliability of knowledge
by Dave Elder-Vass

Daniel, thank you for a very constructive and accurate post, as always! Rather than picking on some detail to disagree with, I’d like to add something, if I may. In addition to the larger argument you’ve covered in your post, my recent book The Reality of Social Construction seeks to apply this argument to various contentious questions of social theory, and it might be useful to illustrate one of these. Perhaps the example that fits best with your interests is the discussion of the social ontology of knowledge.

One of the big issues in late C20 social theory was generated by poststructuralism’s tendency “to challenge conventional assumptions about knowledge by exposing the dependence of knowledge claims on unacknowledged social influences” (207). But these challenges were knowledge claims themselves, and it was never clear “how they might attain some kind of reflexive equilibrium in which they contribute productively to our understanding of knowledge without undermining their own epistemic status” (207).

I try to resolve this by developing an ontology of knowledge that explores how it might simultaneously be socially constructed and also potentially reliable (though never absolutely indubitably certain) because it is also influenced by the phenomena that it is about.

The relevant chapter argues that knowledge is a kind of authorised belief: beliefs are accepted as being knowledge when they have been formed in accordance with knowledge forming practices that are socially approved (normatively endorsed) for the kind of knowledge concerned. We are prepared, for example, to accept knowledge claims about empirical facts when they are plausibly justified on the basis of observation reports. I know that a black car just went by my window because I saw it, and others would be perfectly willing to accept this claim (unless they had reason to doubt that I really did see it) because observation is considered a suitable knowledge forming practice for claims of this sort: this is an epistemological norm, supported by an epistemological norm circle.

I began with a non-scientific example, as I wish to swim against the tide that sees scientific knowledge as the paradigm case of all knowledge; the assumption that it is may obscure important features of knowledge understood more generally. But scientific knowledge is also an important case, and a complex one, as there are several distinct but interrelated sets of knowledge forming practices implicated in it. A student may claim to know that biological species evolve because she has read this in a suitable textbook, and most of us would be willing to accept this claim because reading from reputable scientific books is considered a suitable knowledge forming practice for claims of this sort. (There is also, in this case, a competing epistemological norm circle that supports the claim that species do not evolve on the grounds that this is an issue where reading from religious books is the appropriate knowledge forming practice). Practising scientists, on the other hand, who are generally engaged with less stable claims than those to be found in the textbooks, take a somewhat different approach. They tend to accept those claims that are endorsed by those individual scientists who are considered authorities in the area concerned. This, too, is a socially authorised knowledge forming practice. At a third level, we have those knowledge forming practices employed by the ‘authorities’ themselves; clearly these vary enormously depending on the object of study, but we may generalise and say that the critical element is that knowledge claims are ultimately to be judged by consistency with the observed empirical facts, although there are reasons, familiar from the work of Kuhn, Quine and others, why this may not be the case in the short term.

Knowledge is therefore dependent on the normative endorsement of specific practices as being appropriate for producing beliefs that may be labelled with the honorific ‘knowledge’. It is thus socially constructed and certainly dependent on social influences. But does that make it unreliable? That depends on how effective the practices concerned are in producing reliable knowledge claims. When the practices are such that their conclusions are strongly influenced by the causal impact of the objects that the resulting belief is about, I suggest they are more likely to produce accurate knowledge.

But knowledge forming practices are not only shaped by such considerations. Most significantly, perhaps, those with social power tend to enforce standards of belief that favour their own legitimacy. Knowledge forming practices in strongly patriarchal societies, for example, may tend to endorse knowledge claims that support the balance of gender power even though they fly in the face of other types of evidence, as numerous feminist writers on science have demonstrated. We are still too close to that kind of society to be complacent about the accuracy of all of our scientific knowledge claims, but at the same time the massive technological development of modern society indicates that in many respects our scientific knowledge-forming practices have been unprecedentedly accurate about the world they seek to explain.

The issue here is not that social influence undermines the reliability of knowledge: all knowledge by its very nature depends on social influence, in the sense that claims only come to be accepted as knowledge if they have been obtained in socially approved ways. But some kinds of knowledge forming practices (and thus some kinds of social influence) may produce more accurate knowledge than others. In strongly differentiated modern societies, there is perhaps space for more accurate knowledge forming practices to develop in some domains, such as the natural sciences and everyday empirical knowledge, even when those in others remain more contestable.

This leaves us with a perspective on knowledge that recognises it is always at risk of being wrong, but also accepts that some knowledge claims may be well founded. This is not a perspective that undermines itself, but it is certainly one that demands humility over the possibility of error.


Anonymous said...

This is a fascinating discussion though I wonder why "knowledge" is the point of discussion rather than "truth". Consider this sentence from the Elder-Vass post: "Knowledge is therefore dependent on the normative endorsement of specific practices as being appropriate for producing beliefs that may be labelled with the honorific ‘knowledge’." But isn't the primary honorific here "true" rather than "knowledge", true being a status which signals that a collection of beliefs are knowledge? The focus here on knowledge seems to cordon the discussion off from what I imagine to be an extensive collection of (seemingly relevant) philosophical positions. One position which jumps to mind is that of Habermas, in particular his "pragmatic realism", which is founded on the assumption that the objective world is the ultimate standard against which claims are to be evaluated, but which is attentive to the role of language in mediating any such claims. It would seem there is sufficient overlap between the positions of Eder-Vass and Habermas to warrant further exploration. Perhaps this appears in the book?

Doug Blum said...

Great discussion. My criticisms center on the following passage, as an encapsulation of your larger argument. As you put it, “Knowledge is ... dependent on the normative endorsement of specific practices as being appropriate for producing beliefs that may be labelled with the honorific ‘knowledge’. It is thus socially constructed and certainly dependent on social influences.”
My problem with this formulation is partly – as Anonymous suggests – that it appears to conflate knowledge with truth. Truth does not depend on social recognition; a given proposition may be true whether or not we recognize it as such. Knowledge, on the other hand, is shared belief or acceptance; it has a social quality. Going from truth to knowledge therefore requires that claims about the former are grounded in broadly accepted frameworks for adjudicating such claims. However, utterly different epistemologies are used to get at such knowledge, depending on one’s ontological orientation.
Within the sacred ontology of a religious community, the socially approved way of knowing is essentially literal. The Word is, by definition, Truth. Resolving uncertainty about truth claims (or advancing knowledge) therefore merely requires consulting the text. On the other hand, from a scientific-empirical standpoint – consistent with critical realism – we would accept or reject truth claims based on purely objective parameters, i.e., whether our findings are consistent with our causal propositions. This is consistent with an ontological orientation in which reality is objectively knowable. From this perspective, scientific knowledge is not “dependent on social influences,” but vice versa.
A related problem has to do with the distinction between norms and purely practical understandings (i.e., scientific guidelines). The principles underlying each are so categorically different that using the term “norm” to describe both unavoidably muddies the waters: it conflates ought with is. In the case of religiously informed beliefs about patriarchy, norms are shared beliefs about moral “oughtness.” In contrast, scientific guidelines are not considered “appropriate” in the same sense of the word – they are considered objectively valid, inasmuch as they are founded on logical principles.
Which of course begs a final question about the ontological status of logic – and in this respect the above conflation echoes a recent debate between you and Margaret Archer in European Journal of Social Theory. In that exchange you argued, “Logical relations … exist in our heads. We share similar understandings of logical relations because we share similar cognitive capacities and we are taught to use them – to reason – in similar ways. Logical relations are themselves ideas, ideas about the relations between other ideas.” Archer, on the other hand, held that “logic – the principles of identity and non-contradiction – is acquired in natural practice and is a predicate of both being able to think at all and thus also of verbal communication. The understanding of logical relations is therefore prior to any teaching act and primitive to the expression of logic as ideas.” I strongly agree with Archer, and would even go a step further: logic is the exercise of an intrinsic cognitive capacity rooted in brain structure, itself resulting from an evolutionary process determined by the objective properties of physical reality. In sum, I'm skeptical about approaching all knowledge as resting on "normative endorsement."
Those disagreements aside, I thought your book was in many ways absolutely terrific. I’m very persuaded (as I gather Dan Little is also) by the thrust or your argument reconciling the habitus with reflexivity. I had come to an essentially similar although na├»ve conclusion on my own, but I really appreciated the depth and clarity of your exposition. I’m drawing from it in my current manuscript on “The Social Process of Cultural Globalization.”

Dave Elder-Vass said...

Thanks to you both for your perceptive comments. In the book I do distinguish between knowledge and truth. While I agree that 'true' is another honorific used to label particular beliefs as authorised, it also has a distinct meaning from 'knowledge'. We can be relativists about knowledge without being relativists about truth. As I say in the book "A relativist about knowledge... may hold onto the belief that all true ways of describing the world would be consistent with each other and correspond to a single reality that exists independently of our descriptions of it. The realist separates knowledge and truth partly so that knowledge can be treated as socially contingent, while truth remains independent of historical specificities in systems of belief" (231). But that, of course, goes along with recognising that all truth *claims* are fallible.

Doug, I wonder whether logic and morality should both be approached in similar ways. We do have some basic biologically evolved logical capacities, but the ways in which we develop those are strongly shaped by our experience of the world and by the traditions of thinking about thinking that we are socialised into. Similarly, we seem to have some basic biologically evolved moral instincts but the ways in which we develop those are strongly shaped by our experience of the normative environment we are socialised into. Some of our norms get linked strongly into our moral sense whereas others don't, but the normative environment influences them all.

Michael E. Smith said...

Sorry to weigh in late here. In archaeology we discuss these issues in terms of whether we discover or construct the past. Constructionism remains popular among many archaeologists. The example I use in my graduate classes is from Martin Rudwick's The Great Devonian Controversy (1985, Univ. Chicago Press).

“Bookish people with no practical experience of mapping often assume that a map is an unproblematic replica of reality, or merely a miniaturized version of what one would see from the air. Those who make intensive use of cartography know on the contrary that any map is a pervasively conventional representation. They also know that in indefinite number of different maps of the same area can be made for different purposes, yet all may be equally valid representations of the same natural reality."

“To put the point another way, neither ‘discovery’ nor ‘construction’ is by itself an adequate metaphor for the production of scientific knowledge. The outcome of research is neither the unproblematic disclosure of the natural world nor a mere artifact of social negotiation. The metaphor of shaping—or, in the original sense of the term, forging—has been used allusively throughout this book.” (p. 454, 456).

And we don't worry much about "truth" in archaeology. To quote Indiana Jones:

“Archaeology is about Facts. If you want the Truth, go next door to the Philosophy Department.”