Validity - a tale of disillusionment

By Thane on Mon 21 November 2011

In digging around for different people's understandings of social constructionist philosophy for my thesis, I discovered this great journal article by Steinar Kvale (1995) entitled The social construction of validity. What I was trying to do in my thesis was get back to some of our fundamental assumptions about social phenomena, and in doing so, I was drawn much further down the rabbit hole than I expected: I eventually had to start delving into our understanding about the ways in which we understand.

Correspondence, coherence and pragmatic utility

Kvale writes of a number of perspectives on validity and truth, where the most precise definitions seem to contain some measure of correspondence, coherence and pragmatic utility. Traditionally, valid or true statements exhibit all three.

  • Correspondence - does the statement/knowledge correspond to objective reality? (Think: realism, positivism, pretty much most of modern science).
  • Coherence - does the statement/knowledge have some sort of logical, internal consistency? (Think: mathematics).
  • Pragmatic utility - can we actually use the statement/knowledge in some practical way? (Think: pragmatism).

Especially from a postmodernist point of view, saying that a statement corresponds to reality is highly dubious. (See this lecture by Rick Roderick on Derrida's work). The view that correspondence is a real possibility is known as a positivist view, which says that reality can be known directly. For most people's everyday practical purposes, this is certainly not a bad world view - people seem to have gotten by just fine on this view for ages.

It's only when slightly odd people (such as myself) start asking too many questions, drawing them into the details, that this view starts to unravel:

This apple I'm eating is really sweet. I wonder what it's made of?

A variety of different types of cells.

Hmmm, interesting. What are those cells made of?

Molecules, which are made up of atoms.

Cool. But what are atoms made of?

Subatomic particles: protons, neutrons and electrons. And those are most likely made up of quarks.

Wow, I didn't know quarks could be so tasty! What're quarks made of?

We don't really know, but we've got a couple of theories about what they could be made of.

Wait a minute... So what the hell am I eating?

_Good question. Nobody really knows. Like I said, we've got some theories... _

Reductionism and correspondence

The reductionists hope that by isolating smaller and smaller bits of things we'll eventually come up with a true description of "objective reality", i.e. correspondence. They obviously haven't read Sapolsky and Balt's paper entitled Reductionism and Variability in Data: a Meta-Analysis (published in Perspectives in Biology and Medicine, 39(2), 1996 - I had to personally contact Professor Sapolsky to get hold of a copy of it).

One of the major assumptions in reductionist philosophy is that "noise" or variability in experimental data will vanish as we get closer and closer to "objective reality" by examining finer and finer levels detail. In Sapolsky and Balt's paper, they test this assumption: they gathered many years' papers on the subject of testosterone and aggression and organised them into six "levels": organismal, organ systems, single organs, multicellular, single- cellular, and sub-cellular papers. They then catalogued the variability in each and every paper, plotting the six levels' variability against each other.

The expectation: a decreasing trend in variation as they zoomed in more.

The result: no observable trend.

The interpretation: reductionism is helpful, but only to a certain extent. At this point, your options pretty much include moving on to chaos/complexity theory. You've come too far down the rabbit hole to go back to positivist thinking.


For many practical purposes, however, we can talk of apples and the effects of testosterone on primate behaviour, but practical purposes _are something quite different to _correspondence. As for coherence, that's a topic for another day, because I'd have to convince you that mathematics, similar to English, is just another language, rooted in and emergent from interaction between people in society (see this blog article).

Going forward, I've given up on being able to expound "what things are", and especially on being able to write "truth" down or speak about it. Instead, I'm focusing much of my energy on discovering/coming up with relatively stable guiding narratives (knowledge, statements, etc.) that I will test and put to practical use in different contexts, tweaking the narratives depending on the desired results - my personal view of what research should be.

We can't hold on to the promise that someone's going to give us the ultimate answer. I think we've each got to seek it out for ourselves, relentlessly and carefully questioning our own and others' internal assumptions.