Theories and Stories

Abstract

We would like to say from outset that our comments come from position sympathetic to that of McAdams's. Each of us teaches at least one course from an essentially narrative point of view (Developmental Aspects of Folk Tales and Psychological Perspectives on Comparative Mythology). In addition, each of us has extensive experience in traditional natural science methodology of psychology. The questions McAdams wrestles with, we wrestle with, and our comments are, for most part, less criticisms than they are statements of issues that continue to be problematic. An initial issue concerns extent to which one takes seriously, and does something about, contextualization. The problem with focusing on individual narratives of the contextualized adult is that this is unlikely to produce generalizations (laws) describing human behavior, unless individual tales are to be used inductively as reports or observational data for hypothesis formation. And these hypotheses are likely to be interactional ones based on what we learn from persons and contexts. Doing this suggests not that our theories are inadequate or insufficiently tested, but that in 100 or so years of psychology as discipline, our observations have been insufficient. Perhaps they have been. They are limited as any observations must be by theories, implicit or explicit, that tell us what is worth observing. The subject's own report to us also reflects such choices. Description may be, as McAdams states, more basic than explanation, but every description is also, to some extent, an explanation. No one describes from an atheoretical standpoint; it is just that theory is not typically explicit or complete. Whom one chooses to describe, what aspects of their lives are selected, how descriptions are couched-all of these are guided by some (usually implicit) theory. So now, are we to make better, more phenomenologically pure theories by specifying contextualization of observer? And how far are we to go with this? Gender, social class, sexual orientation, birth cohort, unconscious fantasies, an indefinite taxonomy of situational contexts? What we then end up with is deeply contextualized observer regarding deeply contextualized other, both of whom may have purposes for engaging in interaction that are not consciously available, but are still influential. We may say that it is important to know how understand their lives, but this is always accomplished through filter of how we understand people understanding their own lives. If we consider as valid only momentary expression of particular, then postmodernist playful irony risks degeneration into cynicism and kind of scientific know-nothingism. It is not that postmodernist view is inaccurate; it is just not particularly useful for creating scientifically testable ideas. It is useful for reminding us of limitations of any of our ideas in face of other plausible ones. And here, too, narrative approach has something to offer in emphasizing that a good story (e.g., coherence, internal consistency, persuasiveness) makes for plausibility. However, even good stories escape falsifiability. And, if we care about this criterion (the crucial one that distinguishes scientific paradigmatic from narrative approach), then we must invoke standards other than goodness of story. In light of our language becoming more and more scrutinized for its sexual and cultural biases and our research expected to become more and more contextualized, new Puritanism casts its shadow. Science is, by its nature, approximate. No one knows precisely when Brooklyn Bridge, left unrepaired, would fall down. We make informed guesses-about its demise and about construction of better bridges. No one knows for certain if or when their patient will commit suicide. It would be truly tragic if we ceased or delayed research on suicide risk until we could scrupulously contextualize every aspect of ourselves and of our research participants. We work, then, in terms of guiding, approximate, and falsifiable paradigms.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 4 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.