One of the important insights in Bateson's Mind and Nature is that interactions between individuals are of a different logical type than the individuals themselves. Groping with the way to examine relationships of this logical type is one of the major concerns of the book.
Related to this topic, Bateson discusses the negotiated status of "contexts" of interactions. Essentially, both (or all) participants in an interaction must communicate about the framing context of those interactions:
Learning the contexts of life is a matter that has to be discussed, not internally, but as a matter of the external relationship between two creatures. And relationship is always a product of double description.
It is correct (and a great improvement) to begin to think of the two parties to the interaction as two eyes, each giving a monocular view of what goes on and, together, giving a binocular view in depth. This double view is the relationship.
Relationship is not internal to the single person. It is nonsense to talk about "dependency" of "aggressiveness" or "pride," and so on. All such words have their roots in what happens between persons, not in some something-or-other inside a person (2002:124).
I like this sentiment, and Bateson talks about it in reference to his idea of so-called "dormitive" explanations. Of "dormitive" explanations, Bateson (2002:80, emphasis in original):
A common form of empty explanation is the appeal to what I have called "dormitive principles," borrowing the word dormitive from Molière. There is a coda in dog Latin to Molière's Le Malade Imaginaire, and in this coda, we see on the stage a medieval oral doctoral examination. The examiners ask the candidate why opium puts people to sleep. The candidate triumphantly answers, "Because, learned doctors, it contains a dormitive principle.
A better answer to the doctors' question would involve, not the opium alone, but a relationship between the opium and the people. In other words, the dormitive explanation actually falsifies the true facts of the case but what is, I believe, important is that dormitive explanations still permit abduction. Having enunciated a generality that opium contains a dormitive principle, it is then possible to use this type of phrasing for a very large number of other phenomena. We can say, for example, that adrenaline contains an enlivening principle and reserpine a tranquilizing principle. This will give us, albeit inaccurately and epistemologically unacceptably, handles with which to grab at a very large number of phenomena that appear to be formally comparable. And, indeed, they are formally comparable to this extent, that invoking a principle inside one component is in fact the error that is made in every one of these cases.
One might add phlogiston -- the error being that the similarity among many different substances that burn can be explained by the possession of each of these substances of a common essence that causes burning.
A "dormitive" explanation accounts for some aspects of a relationship by means of unobservable "qualities" of individuals. In other words, it is a non sequitur.
The point is that "pride" as an internal quality is nothing more than a folk psychological attribution of a pattern of interactions between two (or more) individuals.
On the other hand, "pride" is also a description of an internal emotional state. This emotional state is defined in the context of a pattern of interactions, but it certainly is a state that genetic or environmental changes might alter. In other words, it is a phenotype. Aggressiveness is also a phenotype, insofar as it can be measured through patterns of interactions or emotional assessments. And it is at least possible that aggressiveness might vary among individuals in ways related to genetic variation.
In that context, saying that opium has a "dormitive principle" certainly does not illuminate the mechanism by which opium induces sleep, but it may serve as a valid description of opium relative to other substances. When administered in appropriate doses, opium is more dormitive than adrenaline. Likewise a person might be described as more aggressive than another, without that description necessitating an aggressive pattern of behavior in any particular case.
Bateson's point is that an adjective should not be confused for a cause.
Only if you hold tight to the primacy and priority of relationship can you avoid dormitive explanations. The opium does not contain a dormitive principle, and the man does not contain an aggressive instinct.
The New Guinea material and much that has come later, taught me that I will get nowhere by explaining prideful behavior, for example, by refering to an individual's "pride." Nor can you explain aggression by referring to instinctive (or even learned) "aggressiveness." Such an explanation, which shifts attention from the interpersonal field to a factitious inner tendency, principle, instinct, or whatnot, is, I suggest, a very great nonsense which only hides the real questions....
I note, that these explanations are precisely the stuff of classical drama -- Molière had the insight to mock the application of such explanations to substances, but what, after all, is Shakespeare's explanation for King Lear's downfall but "pride", or Macbeth's but "ambition"?
...If you want to talk about, way, "pride," you must talk about two persons or two groups and what happens between them....
...All characterological adjectives are to be reduced or expanded to derive their definitions from patterns of interchange, i.e., from combinations of double description.
Bateson applies this method of description briefly to the problem of teaching and learning. The basic idea is that contextual information must condition our consideration of teachers and learners. I found this taken into a logical example in a paper by Edward Redish considering the effectiveness of education in physics:
When we enter a new classroom situation as an instructor, we may inherit environments and constraints that send metamessages that encourage students to frame the class in a particular way. For example, when students arrives [sic] in the classroom shown on the left in figure 17 [a typical lecture room with rows of seats], they tend to interpret the layout as a clear metamessage about what frame to activate and in which to interpret subsequent messages. Most do not expect to interact with friends or the lecturer, most expect to take notes, few expect to think about what is being said carefully and to try to understand. In the classroom shown at the right [with chairs arranged in small groups around circular tables], even on the first day, students will be aware that this is not a traditional classroom. They may activate a group-learning epistemic resource or a new situation resource but they are unlikely to expect a lecture.
Once students have framed a situation, depending on the breadth of their experience and the consequent robustness of their framing, they may have difficulty interpreting overt messages that violate that framing. When substituting for one of my colleagues in a large lecture class, I often tell the students that I plan to have activities that will require student engagement including thinking, evaluating, and stating their views in public. Very few students take me at my word. When I call for a vote on a question, typically only about half the students respond. It is only when I call on one of the non-respondents and ask him to explain why he was unable to decide on an answer that the students begin to take me seriously (Redish 2004:32-33).
In that example, the relationship of teacher and students, and their behaviors relative to each other, are defined in part by the space that they occupy. The teacher can modify that relationship only by attending to the qualities of the interaction -- in this instance, by violating student expectations. That intervention requires the recognition (whether explicit or not) of the characteristics of the space that set the contextual frame for the interaction.
I am saying that there is a learning of context, a learning that is different from what the experimenters see. And that this learning of context springs out of a species of double description which goes with relationship and interaction. Moreover, like all themes of contextual learning, these themes of relationship are self-validating. Pride feeds on admiration. But because the admiration is conditional -- and the proud man fears the contempt of the other -- it follows that there is nothing which the other can do to diminish the pride. If he shows contempt, he equally reinforces the pride.
Similarly, we can expect self-validation in other examples of the same logical typing. Exploration, play, crime, and the Type A behavior of the psychosomatic studies of hypertension are equally difficult to extinguish. Of course, all these are not categories of behavior, they are categories of contextual organization of behavior (Bateson 2002:126).
This follows a description of play between Bateson's dog and gibbon, in which patterns of interactions spontaneously emerged and repeated. The idea of external context framing interactions between individuals, and individuals reacting to external context and social contexts iteratively is very compelling.
Bateson G. 2002. Mind and Nature: A Necessary Unity. Hampton Press, Cresskill NJ. Amazon
Redish EF. 2004. A theoretical framework for physics education research: modeling student thinking. In Proceedings of the Enrico Fermi Summer School in Physics, Course CLVI. Italian Physical Society.