"A Difference That Makes A Difference"

Over the years, I've heard people, especially those involved with NLP (Neuro-Linguistic Programming), use the phrase 'a difference that makes a difference'. It has a catchy sound to it and the notion intrigues a lot of people who, at the very least, think of it as a neat kind of play on words. Where did the usage come from? 

Difference and Differences That Make A Difference
Difference qualifies as one of the most important fundamental undefined terms in the non-aristotelian worldview that Alfred Korzybski  delineated. In a world where the premise of non-identity rules (the world we actually seem to live in), where no two objects however similar are 'exactly the same in all respects', difference indeed seems fundamental. 

As a term, difference, also qualifies as one of the important multiordinal terms that Korzybski talked about. Although, I can give a dictionary definition of a multiordinal term, the specific 'meaning' of a multiordinal term only becomes clear in a specific context of discussion when one specifies the level of abstraction the word refers to. What does that 'mean'? For now, I'll roughly give Korzybski's simple rule for determining if a word qualifies as a multiordinal term: Can the term be applied to itself and still make  sense? Can you fear fear? Can you be in love with love? Can you hate hate? Can you feel anxious about your anxiety? Can you have knowledge about knowledge? Can you think about your thinking?  This is not just playing with words, but points to the multiordinal (multi-leveled) and concomitant self-reflexive character of the human nervous system and human evaluating. Thinking2 about thinking1 is not the same as thinking1. And how you think about what you think may be crucial. 

Okay, back to the subject of this little (still) article: difference, another multi-ordinal term. In a non-identity world, a world of differences, we can have differences that make a difference (for given individuals with given purposes) and differences that don't (when we recognize similarities). For sane evaluating we need to recognize both similarities (where the differences among the different individuals viewed as 'same' don't make a difference for our purpose) and differences that do make a difference. 

For Korzybski, difference was not the absence of 'sameness' (that would make 'sameness' fundamental, which seems a basic structural assumption of the aristotelian metaphysical worldview). Korzybski reversed this: 'sameness' or more clearly similarity constituted a species of difference; Difference, not 'sameness', seems more basic. As he also pointed out, a world of absolute difference between two things seems as impossible as one where absolute sameness exists. With every thing entirely different from everything else, recognition and knowledge would be impossible. In a non-identity world, difference appears fundamental; there exist only different kinds and degrees of differences. (See Science and Sanity, page 165.) What I've written here may—or may not)— seem obvious; Korzybski wanted to get people to understand it and, more importantly, apply it in their everyday evaluating (not necessarily so easy, even if understood). 

Although Korzybski may have used the phrase 'a difference which makes a difference' somewhere, I have thoroughly searched through his published writings and have not found it used. Korzybski's student and co-worker Wendell Johnson seems to deserve credit as the first to specifically talk about differences that make a difference and those that don't. Here's a bit from the discussion in his 1946 introduction to GS, People in Quandaries:

The law of identity sometimes holds sufficiently for practical purposes, in spite of its structural defectiveness. Therefore we can use it many times, but we should always be aware of our use of it. When eating peanuts, for example, we may proceed on the practical assumption that peanuts are peanuts, that peanut 1 is peanut 2, that they are the same. Even so, we should remember that they are not, that ultimately peanuts are not peanuts, that is, peanut 1 is not peanut 2. The differences, generally speaking, make no important difference, of course, and we can for the most part disregard them. But if we are basically oriented to non-identity, we will bite into the bad peanut that may be found in almost any bag, without bursting into invective against "these damned peanuts." We will merely discard peanut1 and go on to enjoy peanut2, since basically we had not assumed that they would be the same anyway. Therefore, an occasional bad peanut is not cause of shock, no generator of tensions.

It is "in principle" that different things are the same. This is to say, it is on relatively high levels of abstraction, "in general," "on the average," "for practical purposes," "in main essentials," with regard to certain more or less important respects (not in all respects) that two different things may be evaluated, spoken of, or dealt with as though they were identical. For certain purposes specific differences may not make a difference; what is important is that we realize that differences exist and that we recognize the conditions under which they do make a difference. This is to say that what is important is that we be fundamentally oriented to non-identity so that we shall be prepared for differences at any time. It is precisely the differences we least expect that tend to make the biggest difference, that have the gravest consequences and demand the most difficult readjustments.  [People in Quandaries, pages 178-179]. 

Gregory Bateson and Difference
The usage 'a difference that makes a difference' is usually ascribed to the non-aristotelian (in the korzybskian sense) anthropologist Gregory Bateson, who used it in the Alfred Korzybski Memorial Lecture that he gave for the Institute of General Semantics (IGS) in 1970. Bateson had had a long peripheral interest and association with Korzybski's work, having briefly corresponded with Korzybski in the late 1940s (the two men exchanged papers), and then having written a number of the dialogues with his daughter, he called them "metalogues",  which were published in ETC.: A Review of General Semantics starting in the early 1950s (ETC. was then edited by S. I. Hayakawa and published not by the IGS but by the International Society for General Semantics). 

Bateson certainly qualified as a non-aristotelian before he ever heard about Korzybski (see Naven, a self-reflexive study of the anthropological fieldwork he did in New Guinea among the Iatmul, first published in the 1930s). But at some point, sometime in the 1940s, he appears to have read some of Korzybski's work, and probably other GS literature, including it seems nigh certainly Wendell Johnson's 1946 book, People in Quandaries.

So, as far as I know, as a matter of historical record, Wendell Johnson was the first to use 'a difference that makes a difference' as a significant formulation while explaining Korzybski's premise of non-identity. This takes nothing away from Bateson's brilliance. As a theorist Bateson built upon the notion of difference, and a difference that makes a difference in original and significant ways for the human sciences, starting it seems with his AKML paper: "Form, Substance, and Difference", which I read soon after it was published and which I still recommend most highly. I would find it most instructive for some serious student or students—perhaps they are reading this now—to trace the development of the notion of difference and differences of differences from Korzybski to Wendell Johnson to Bateson in much greater detail than I've done here (without getting lost in verbalism). That would constitute a genuine contribution to korzybskian GS scholarship. Any takers?