Constructivism doesn't have to be skeptical
Lincoln and Guba
A team of evaluators I know1 were conducting a needs assessment for a large government client in which they were tasked with determining the needs in the region. To do this, they gathered lots of interview data from local people about what they thought the needs were, then conducted a good qualitative analysis to code and summarize what they found out. They presented the results of this effort to the government client, hoping that they had identified some gaps. The client responded with indignation: some of the claims made by participants about the resources available were demonstrably false, and now they were in print! For example, some of the community members quoted in the report claimed that certain government programs and resources did not exist in the community, but those programs and resources did exist. Surprised to be encountering any opposition, the evaluators responded that the claims of the participants were “true for the participants.”
Epistemology is an aspect of evaluation theory that seems like a luxury until it suddenly turns into a necessity. In the situation above, some mistakes in epistemology contaminated the methodology of the report and then undermined the credibility of the evaluators beyond repair.
In a theoretical sense, what happened here? The evaluators got confused between several levels of truth:
The participants made certain claims. These were quoted truthfully. ✅
The claims made by participants faithfully represented their understanding of reality. That is, to arrive at a genuine understanding of the participants’ perspectives, the evaluators were right to attend to these remarks. ✅
The referent of the claims made by participants was not true. The participants were wrong about reality. ❌
The evaluators apparently did a nice job of handling levels 1 and 2, and didn’t pay any attention to 3. In our subsequent conversation about this debacle, the evaluators kept emphasizing that the big story was in their discovery that people didn’t believe that resources and programs were available. The beliefs of participants were more important to them than the referents of those beliefs.
Unfortunately, for a needs assessment, we probably need to understand all three levels of the truth here. This the case because different combinations of facts lead to different diagnoses of the problems facing the agency. If people have an unmet need, believe that there are no resources, and they are correct, then the problem is that the agency has a resource gap. If people have an unmet need, believe that there are no resources, and they are wrong, then the agency has a communication problem. These are rather different problems with rather different solutions. (Communications problems are generally cheaper to fix, for one thing.) By mixing up these levels of truth, the evaluators actually generated confusion for the stakeholders instead of clarity.
For most people who aren’t social scientists, the approach of the evaluators in the story above was silly and my diagnosis of the problem was obvious. Nobody actually thinks this way, right?
Lincoln and Guba
Allow me to introduce the radical social constructivists, Yvonna Lincoln and Egon Guba, wife and husband coauthors who are famous in the fields of evaluation and social science. It is basically impossible to get through a survey course on evaluation theory or qualitative methods without reading them. Here is how they describe their epistemology, which they call “naturalism”:
Social reality is not objectively "out there," but exists only as a series of mental and social constructions derived via social interactions. Rather than looking for an external reality the naturalist looks for internal realities—the sense-making and belief structures that order human existence and exist only within individuals. It is the holders of those realities . . . who provide whatever warrant exists. The warrant is thus no more powerful or pervasive than are the persons who hold the constructions.2
So what about things external to the minds of individuals? Should we seek to verify that they are real or true? Under Lincoln and Guba’s view:
The primacy of verification procedures is eliminated since there is nothing to verify. The object of naturalistic inquiry is to identify and describe various emic constructions and place those constructions in touch—with the intent of evolving a more informed and sophisticated construction than any single one of the emic constructions, or the researcher's or evaluator's etic construction, represents.3
We can summarize this position as the claims that: a) social reality only exists internally to individuals, b) the warrants for claims about this reality are found only in the mind, and c) it is impossible to verify these claims about social reality.
Alright, this was from an evaluation book from the 80’s, which were kind of a crazy time for philosophy and the social sciences.4 Maybe they updated their view? Here’s Lincoln and Guba in 2013:
There is no compelling reason to believe, a priori, that this surround [i.e. the world we perceive] has any existence apart from the individuals who encounter it, that is, to believe it to be objectively independent of the sense mechanisms of the individuals who experience it.5
Here we find Lincoln and Guba denying the existence of any reasons to believe in a mind-independent reality. This is not just about “social reality”, but about all our experiences of reality. Just in case you don’t get the point, Lincoln and Guba double down a few pages later:
Thus, “reality,” “truth” (including truth viewed as a “regulatory ideal”), and “fact” are all relative concepts—they are themselves semiotic signs that are relative to the person(s) who hold particular sense-makings, constructions, or meanings.6
In terms of our example above, the participants who claimed that specific community resources did not exist were not wrong even if those resources did exist because reality is relative to them. In fact, the assertion that the resources exist is just another relative perspective. There is nothing that we could do to verify that the resources exist. For example, if the resource is an addiction help hotline, it wouldn’t do any good to call the hotline and listen to hear whether anyone picks up. That would just be more data in my mind and would in no way invalidate the claim that the hotline does not exist.
If Lincoln and Guba are correct, then there is nothing wrong with the needs assessment in the story with which I began this essay.
Alternative views
Many people who are attracted to constructivism are drawn by its detailed accounts of how social phenomena occur. How do people learn to perform gender? The constructivists show the many small steps that go into socially constructing manliness and socializing people into it. How does a substance abuse counsellor teach addicts how to change their lives? This involves the building up of an idea of sobriety that is tenable and desirable, then helping people in their process of identity formation as sober, and so forth. These are often compelling accounts that answer important causal questions.
However, constructivism does not necessarily entail the kinds of deeply skeptical epistemology that Lincoln and Guba have advanced. The rejection of the possibility of truth and the denial that it is possible to know anything about reality has some basic problems. First, the fact that we all agree on many aspects of reality would seem to be a complete accident if there is really nothing out there. Second, there is the paradox in advancing that it is true that there are no truths – this is right on par with “this statement is a lie.”
One popular alternative epistemology to skepticism is critical realism. In critical realism, reality is held to exist autonomously from our perceptions, but knowledge is mediated by cultural and scientific practices. While acknowledging epistemic fallibility, critical realism rejects radical doubt and seeks to approximate truth by degrees.
A less popular alternative view, but one that I find very interesting, is the idea that knowledge of the world comes from the exercise of intellectual virtues, like openness and humility. This virtue epistemology is rooted in the qualities of the knower rather than the abstract justification of beliefs via truth procedures.
Coherentist epistemology says that truths can be known by fitting into a web of existing beliefs. There are observations from the sociology of science that would seem to support this idea. Classical hermeneutics and phenomenology are also less skeptical than radical constructivism, since both developed procedures for iteratively learning about the world through interpretation and investigation. Modern fans of both of these schools of thought, including Lincoln and Guba, tend to ignore the realist roots of these schools of thought.
The point
Constructivism is compatible with all of the above epistemologies. We don’t have to be radical skeptics and deny the possibility of learning the truth. Rather, it makes more sense, I argue, to say that we can have genuine knowledge about socially constructed things. Measurement of psychological attributes demands it and much of our common language presupposes it. For example, we can have a conversation about trends in fashion or writing or about how gender roles have changed in the last 100 years. In these conversations, we can say true things and can grant that some people have more knowledge. The signature aspect of constructivism is actually its ontology – the way it argues the world is put together – rather than its epistemology.
I know a lot of evaluators, so there’s no use in you trying to deduce who this story is about. If you are tempted to try, please use your brain for something else today.
Guba, E. G., &. Lincoln, Y. S. (1989). Fourth generation evaluation. Newbury Park, CA: Sage.
Ibid.
A wonderful review of the 80’s and 90’s evaluation theory is found in House, E., & Howe, K. R. (1999). Values in evaluation and social research. Sage Publications.
Lincoln, Y. S., & Guba, E. G. (2013). The Constructivist Credo. Left Coast Press.
Ibid.