I just now mounted a transcription of MS 403 (1893), "The Categories", at Arisbe.
http://members.door.net/arisbe/menu/library/bycsp/ms403/ms403.pdf This is a rewrite -- up to a point -- of the 1867 paper on the categories, and I include in the transcription of the later paper a copy of the 1867 paper interleaved with it in such a way as to make it easy to compare the two as regards what is and is not changed. The changes are, in general, explainable in terms of the different audiences for which they are written. In the case of the later paper, the audience would be the reader of a logic text in which it (MS 403) was to appear as Chapter 1. The name of the logic text (never published) was to be "The Art of Reasoning" and -- judging from the name -- it seems to have been intended for people of the same type as those whom he recruited for his "distance education" course ("correspondence" course) in the late 1880's since that was also the advertised name of his course. (See Nathan Houser's account of this remarkable endeavor in Volume 6 of the Writings of CSP, Indiana University Press). I don't know whether Peirce was still thinking in terms of that correspondence course in 1893, which seems to be several years after he gave up on the course; but the reference to logic as an art rather than a science and the use of the word "reasoning" rather than, say, "reason", suggests that the intended readership was the same, adults primarily concerned with what logic could do for them as good thinkers generally. Needless to say, perhaps, Peirce's idea of what would appeal to those interested primarily in practice rather than theory seems a bit odd and unrealistic at times. But watching "Deadwood" has convinced me that Americans may well have tended to think about things in a more eloquent and intelligent way in those days than we are presently accustomed or inclined toward nowadays -- an idea which has occurred to me a number of times in the past when reading not only Peirce but some other American writers of the late 19th Century -- so maybe Peirce wasn't so far off in his expectations about his prospective students as we are inclined to think. Anyway, I find the modifications Peirce did and did not make in his 1893 rewrite of the New List helpful in understanding his thinking generally and perhaps others will as well. Unfortunately, MS 403 stops just one sentence short of the passage in the New List where he defines the symbol in terms of "imputed quality", though he has just drawn the distinction between an internal quality and a relative quality, as in the New List but does not complete that with the notion of the imputed quality nor make use of the talk of three kinds of quality to define the icon/index/symbol distinction. The reason seems fairly clear when we turn to MS 404, which was apparently composed as a continuation of 403 but introduces something for which there is no corresponding passage in the New List, namely, an attempt at a loose, suggestive, intuitive, poetic appreciation of the three-category conception. One obvious reason is that he could not reasonably expect someone who is reading a book on the art of reasoning to understand what is happening in distinguishing between internal, relative, and imputed quality. I do not think it was because he had abandoned the earlier idea of the symbol as being grounded in an imputed quality, since this is really the same as to say that the proper interpretant of a symbol interprets it as if it were an icon conventionally associated with the symbol which is being indexed by the symbol replica. (This is his later doctrine, stated again and again by him from the 1890's on, ) But I don't think it is only that he had decided on a better way of saying the same thing, but also had something to do with the distinction between three kinds of quality: roughly, monadic, dyadic, and triadic (i.e.internal, relative, and imputed quality). What is problematic in this is that in order to make sense of that distinction he had to distinguish between the firstness of firstness itself and the firstness of secondness and the firstness of thirdness since the quality could not otherwise iconize existential or dyadic relations and three-term representation relations. He does of course recognize THAT distinction later, but that is a complication that he would not want to be burdened with explaining in an introductory text in logic! Anyway, I doubt that he had realized the necessity for that when writing the 1867 paper, but I see no reason why it should be thought of as inconsistent with it. All that is required to recognize the foundational character of the New List for his later as well as his earlier work is to be able to understand it as consistent with such further developments of it as turned out later to be required. Nobody holds -- so far as I know -- that Peirce's thought did not DEVELOP across his lifetime: the important question is what can and cannot be regarded as consistent with and/or implicit in the original version, allowing for the need for minor modifications here and there. What people who want to ignore the New List as impertinent are claiming is that he abandoned this or that important part of it, such as, for example, the trichotomy of "likeness"/ index, and symbol as being established there (contrary to Peirce's own later view that there is no radical change in that trichotomy in later years). Attention has usually focused on a supposed "revolutionary" new view of the index subsequent to the l860's: a view inherited from Murray Murphey. But equally as important is the question whether his later doctrine of the symbol as properly and essentially interpretable as if it were a certain icon associated with it was already in the New List in the definition of the symbol in terms of it being grounded in an imputed quality. Short's basic criticism of Peirce's view is based on an account of it which simply ignores the definition of symbolism in the New List, and expressed in the claim that the meaning of a symbol is infinitely deferred owing to symbol necessarily being interpreted AS symbols, which supposedly means that meaning is never achieved at all, whereas the plain meaning of the view that symbols are grounded in imputed quality is that the symbol is not interpreted as such at all if it is not given content by the "as if" interpretation of it as an icon. (The issue of the "pure" symbol is all about this, too, as in Awbrey's claim that symbols can do what they do as symbols without benefit of iconization in interpretation.) So, the point here is that MS 403 doesn't settle anything about the matter of the imputed quality since it breaks off just before it is mentioned and the continuation in MS 404 doesn't rely upon that but simply assumes that one is to understand it is in virtue of this icon association that the symbol has its distinctive meaning. Read closely, MS 404 can be seen to be taking it for granted that the conventional association is not of a symbol with another symbol -- why would he have thought that this could provide the symbol with its distinctive meaning? -- but rather with a quality such as could be introduced only by an icon. That is the basis for pragmatism, which is already there in the New List. Anyway, the URL for the version of MS 403, called by Peirce "The Categories", is http://members.door.net/arisbe/menu/library/bycsp/ms403/ms403.pdf I want to make it available in HTML as well but that will take me a few days. I also want to put up a combination of MS 403 and 404 as a single document which I am currently also at work on editing, but that will take longer. But MS 403 is already available both at Arisbe and at the Peirce Edition Project website as "What is a Sign?" -- Internal Virus Database is out-of-date. Checked by AVG Free Edition. Version: 7.1.394 / Virus Database: 268.10.0/388 - Release Date: 7/13/2006 --- Message from peirce-l forum to subscriber archive@mail-archive.com