In a recent article, Kenneth Gergen (2018) offers a summary of his work on (and against) empirical research in psychology. The article is clearly written, and there are many positive things one could say about it. I am going to focus primarily on points with which I disagree. Gergen and I share common “enemies”, but I am not yet as pessimistic as he is with regard to psychology as a scientific (truth-seeking) enterprise.
We find in this article the outline of two contradictory positions, both of which are defended by the author. First, Gergen argues, because of the tautological nature of psychological statements, our explanations do not escape what is already necessitated by language. If we think carefully about an explanation that makes sense (e.g., John ate the cake because he was hungry; Mary smiled at Beth because Mary is friendly), we realize that the explanation does not–and cannot–offer anything new in terms of psychological principles. We do not discover anything new about the meaning of hunger or how hungry people act. Understanding the meaning of hunger involves understanding the type of action people take when they have access to food; similarly, understanding the meaning of friendliness entails understanding the type of behavioral dispositions displayed by a friendly person.
Thus, one makes psychological sense by explaining a given behavior in terms of a “miniaturized” form of itself.
Gergen (2018, p. 697)
I think there is, unfortunately, a serious blunder here. The relation between an action and a psychological concept that explains the action is not a relation between a thing and its miniaturized version (satisfying your hunger is not a miniaturized form of eating a cake). The relation is between a subordinate/concrete concept (eating a cake) and a superordinate/abstract concept (satisfying one’s hunger). When we describe someone as a friendly person, we are organizing a large set of actual and potential actions with the help of a superordinate concept. The relation between superordinate concepts and their subordinate instantiations is not a tautological relation. It is a conceptual relation that is shaped by (and further shapes) experience.
To say, ‘Mary was friendly to Beth’, is to say something that is at the same time more and less than the alternative statement, ‘Mary smiled at Beth’. Describing Mary’s behavior as friendly expresses a larger set of possible actions (more), while at the same time being under-specified (less), compared to describing her smiling at Beth. By equating subordinate and superordinate concepts and propositions, Gergen permits himself to reduce all psychological statements to tautologies. In a sense, he is correct to point out the semantic (axiomatic) relation between propositions at different levels of abstraction, but information is lost if we project everything onto a single level of abstraction, which is to say propositions at one level of abstraction do not render corresponding propositions at other levels redundant.
In addition to the argument about linguistic necessity, Gergen also offers an argument about linguistic indeterminacy.
[B]ecause each definition of a mental term relies on another mental term for its meaning, we enter a condition of unbridled diffusion of definition. […] Through extended definitional sequence, we find that any given behavior can be explained by virtually any randomly drawn motive or trait.
Gergen (2018, p. 697)
If linguistic determinism and the idea of pseudo-empirical research doesn’t persuade you to drop psychological research, you should consider the complete opposite problem–linguistic indeterminacy. If someone says: ‘Mary smiled at Beth because Mary is friendly’, we find the statement to be intelligible. But what if someone tells you, ‘Mary smiled at Beth because Mary is ambitious’? Is the latter statement nonsense? Here, Gergen correctly points out that we can fill the missing detail of the statement such that the explanation (ambition of Mary) can be meaningfully connected to behavior (smiling of Mary). We could, for example, imagine that Beth is Mary’s co-worker and that Mary wishes to develop good relation with co-workers as part of her career-related ambitions, which is why she smiled at Beth.
What if we read, ‘Mary smiled at Beth because Mary isn’t a very sociable person’? We can still find such a statement intelligible, with the help of what Gergen calls a self-repairing maneuver. Self-repairing is the assumption that people generally wish to change their undesirable psychological states or traits, including the trait of being non-sociable. Mary smiles at Beth because she doesn’t wish to remain non-sociable. That is, someone’s non-sociability can be taken as an explanation for their smile, just as the opposite (their sociability) can explain their smile. This is what Gergen means when he writes, “any given behavior can be explained by virtually any randomly drawn motive or trait.”
The statement is, I believe, a deepity. Of course, it is true that a sufficiently ambiguous and under-specified statement like ‘Mary smiled at Beth’ is open to a wide variety of interpretations. We can find among those interpretations opposite pairs (e.g., Mary is sociable and Mary is non-sociable). It is, however, false that those opposite interpretations will be equally valid and true in the case of an actual, concrete situation involving a Mary, a Beth, and a smile. For a concrete instance of Mary smiling at Beth, Mary’s action cannot be explained by “virtually any randomly drawn motive or trait.”
Contrary to Gergen’s position, and despite the BIG names he keeps dropping (Wittgenstein, Saussure, Derrida, Kuhn), participating in language and culture doesn’t render psychological statements necessary or superfluous, especially when those statements are made with reference to actual and concrete situations. What is special about actual situations is their openness to exploration and further specification (dare we say it, empirical research).
Reference
Gergen, K. J. (2018). The limits of language as the limits of psychological explanation. Theory & Psychology, 28 (6), 697-711.