{l.2} Cue

A. The Notation. The closest conceptual equivalent to a cue is designation [5.3.1] which, as shown by the notation scheme copied in Schedule II, embraces three subordinate concepts: symbols, [5.3.1.1] terms, [5.3.1.2] and names, [5.3.1.3]. Let me delay a discussion of "symbols" and "terms" until later -- under {1.2.1} tag; and {1.2.2} notation. Here, however, a comment on names seems appropriate because it has no counterpart in Onomantics -- rather, it belongs to the parallel field of Onomastics, the other branch of Onomasiology.

The treatment of "names" has changed since the publication of the 1990 standard version of ISO 1087. In that version, names were defined as identifiers of individual objects, and concepts were defined as notions based on the properties of two or more objects. This meant that names could not be designations without contradicting the definition of this word offered in [5.3.1] --see Schedule II.

In overcome this difficulty, perhaps, the notion of individual concepts was introduced in subsequent revisions -- e.g. in the 1994 version listed below. It was defined as a concept "that refers to an individual object," and "Saturn" is mentioned as an example -- the word, 'Saturn,' is viewed as a human construction used to represent the "concept" of a planet that moves around the sun. To support this usage, it is argued that the object itself has no name, but we have a "concept" of the planet, and that is an individual concept.

By contrast, general concepts are defined as those that relate to two or more objects, replacing the 1990 definition of concept, which was then re-defined to mean a broader concept, a notion based on one or more objects, not two or more objects. No doubt there are philosophical reasons for making the distinction between an object and its individual concept. However, since Onomastics provides theories to explain the naming of objects (individual concepts) and Onomantics deals only with the designation of general concepts, the onomantic notion of a cue limits this concept to expressions that represent general concepts. On the premise that, in Onomantics, there is no need to discuss individual concepts, the word "concept" can be used unambiguously in this context to mean only "general concepts".

Does Terminology, by contrast, need to make the distinction between individual and general concepts? Actually, I believe, all the examples of designation found in ISO 1087 refer to general concepts, not to objects (or individual concepts). Moreover, so far as I can see, terminologists typically focus on general, not individual concepts⁵.

If so, are the philosophical considerations that lead to this distinction really needed in Terminology? I cannot be sure the answer is "no," but if one considers the Lexicographic origin of many terms borrowed in Terminology, one can easily find an explanation for the inclusion of name in ISO 1087. I suspect that when terms were originally selected for inclusion in the glossary, 'names' were bunched with 'symbols' and 'terms' without much analysis of their meanings -- the treatment of names in dictionaries raises difficult problems for lexicographers, as explained above, see p.3 ?? . The subject is also discussed in works on Semantics⁶.

Later, when it was decided to subsume these three terms as subordinates of designation [5.3.1], an unexpected paradox emerged. If designations refer only to concepts, how could they include the names of objects if concepts, by definition, were based on the properties of two or more objects. One solution to this difficulty could have involved removing 'name' from the list of types of "designation". The record for [5.3.1.3] could have been deleted -- or a new notation could have been assigned that would associate "names" with "objects" rather than with "concepts".

By contrast, the introduction of "individual concepts" made it necessary to revise the definition of "concept" to make it include the "properties of a set of one or more objects" (draft of ISO 1087, 1994). The new distinction between individual and general concepts, therefore, not only broadened the concept of concept but it made the word equivocal ⁷ in the sense that it now has two meanings for terminologists -- its original meaning (as prescribed in 1990) and a new meaning (in the revisions). Undoubtedly some ISO members will think of the first definition of a concept (including only general concepts) and others of the second (including both individual and general concepts), thereby increasing the potential for ambiguity when this key word is used⁸.

To summarize, if an object can be viewed as a concept (individual concept) then the meaning of "representation" is also broader than that of "cue" because it includes the names of objects as well as expressions used to represent "general concepts".

B. The Description. Turning from their notations, let us compare the following concept descriptions for cue and designation:

[5.3.1] designation: Any representation of a concept [3.1]. {1.2} a representation that succinctly identifies a particular concept: cue:

{1.2} excludes concept descriptions -- they are not "succinct," in the sense that a lexeme is succinct (a minimal form that conveys a meaning) by contrast with texts (containing two or more lexemes). In context, as shown above in A., all "designations," as understood by users of ISO 1087 are, indeed, "cues," but the defining text in [5.3.1] could well be understood as broad enough to include definitions also. If this text were revised to exclude concept descriptions (definitions), then it would identify something that closely resembles a cue {1.2}.

However, there is another problem. The definitions of {1.1} and [5.3.1] are not equivalent if different concepts are intended by the use of "concept". If this word is defined, as it was in 1990 [3.1] -- "a unit of thought constituted through abstraction on the basis of properties common to a set of objects [2.1]" -- then the onomantic and terminological senses of "designation" and "cue" would be quite similar, but when "concept" is revised to include the properties of a single object, then "designation" and "cue" become more different. Actually, concept designates the most fundamental concept in Onomantics: in an earlier and longer draft of this article, I discussed the concept of a "concept," but in order to limit the length of this essay, I have eliminated this material, reserving it later treatment.

Nevertheless, we should be clear about the effect revisions of a "definition" have on concepts. In a semantic context, it is easy to assume that words have meanings which have been poorly defined -- if so, revisions may clarify these meanings. A new dictionary, for example, may claim that its definitions of established words are better than those of a rival, or that a new edition improves on definitions found in an earlier version.

In an onomantic perspective, however, words are only tags to represent ideas that have been described in a text which identifies the essential characteristics of a concept. Reverting to the previous section, we can see that
(1) an idea based on "all the properties of a single object" is, clearly, different from
(2) an idea based on "properties shared by two or more objects".

These two notions suggest a superordinate concept:
(3) includes both (1) and (2).

In the 1990 text of ISO 1087, "concept" was used to represent (2), but in 1994 it had been replaced as the designator of concept (3). Two new terms were then needed for (1) "individual concept" and (2) "general concept". In this simple example, the word 'concept' became a metaphor as its meaning was changed from (1) to (3), and users of the glossary who could not remember its new meaning find the word has become equivocal for them. The meaning of a word was not clarified by the revision -- actually, its meanings had become more confusing.

C. The Tag. Let us now look at the tag, designation, chosen to represent concepts [5.3.1]. Semantically speaking, to designate is a polyseme: among several other senses, it may mean either the signification of a concept, or the name of an object, but not both at the same time. In [5.3.1], "designate" was used only in the first of these senses, i.e. to represent concepts derived from two or more objects. However, in ordinary language, one may say that a person (one object) has a "designation" just as one may speak imprecisely of "naming" a concept.

Consequently, "name" and "designation" are, semantically speaking, polysemes that sometimes refer to concepts and sometimes to objects -- each word, therefore, can represent two different concepts needed by terminologists, which makes them equivocal. To overcome the resulting confusion, one may coin neologisms, such as nom and tag -- as explained in note no.8. Although "tag" admittedly has other meanings, it can be used unambiguously in Onomantics to signify any cue that represents a (general) concept (but not an object, or an "individual concept") and "nom" can be used to refer unequivocally to the names of objects (individual concepts).

Because most people resist neologisms -- a subject I will discuss in more detail in Part III of this article -- it must have seemed preferable to the authors of ISO 1087 to recommend familiar words rather than neologisms: since "designation" usually refers to concepts rather than to objects, and because "name" typically identifies a single object instead of a concepts, it seemed reasonable to stipulate just one meaning for each of these words. By this stipulation, they hoped, apparently, that these two polysemes could become monosemes -- at least in the context of Terminology.

However, when they realized that they had (perhaps unintentionally) included name [5.3.1.3] as a type of designation [5.3.1] for concepts (based on two or more objects), they decided to resolve this paradox by expanding the meaning of concept [3.1] to include not only concepts based on one two or more objects, but also those based on one -- i.e., the names selected for them. In Onomantics, the best way to solve the semantic problem generated by the polysemy of words like 'name' and 'designation' involves finding unequivocal terms for each necessary concept, even if this means coining neologisms (see note #8). The alternative path of stretching fundamental notions like that of a concept will surely, I think, only compound the difficulty.

Incidentally, lexicographers themselves are quite clear about this matter and they make a sharp distinction between concepts and objects -- see the discussion of "terms" vs. "names" following Schedule II above. Let me amplify that discussion by noting that in the Oxford English Dictionary one may find "London" as part of some fixed phrases (e.g., london clay, london lady, london purple, etc.) and in coinages like 'londonesque' or 'londonize,' but there is no separate entry for London as the name of a city. By contrast, in Webster's (1991) [but not W3], names are entered and capitalized, but these entries do not contain definitions of concepts -- instead they provide some information that identifies the persons or places they name.

In general, we classify and use concepts nomothetically in the organization of knowledge as in theories and hypotheses. By contrast, objects are listed, idiographically, and they are found in case studies and descriptive narratives. Textbooks and scientific findings are based on concepts arranged systematically, but telephone directories, gazeteers, Who's Who's and chronologies contain lists of named objects. The former can be developed onomantically, while the latter are analyzed onomastically. The definitions of "concept" and "designation" found in ISO 1087 blur this fundamental distinction.

If terminologists would use designation (tag) to include only terms and symbols that represent general concepts, but exclude names (noms) that identify objects (individual concepts), then Onomantics could abandon "cue" in favor of a more familiar word like "designation". However, if terminologists insist that names are a type of designation and therefore must represent concepts, then we need a different word, like 'cue', to distinguish the designators of (general) concepts from the names of objects⁹.

To summarize, in Onomantics, concepts are represented by descriptions (texts) and cues (short expressions); whereas in Terminology concepts have "definitions" (which, by definition, include links with related concepts) and "designations" (which exclude concept descriptions but include names of objects). I will not say this is "wrong" but I do claim that it produces non- onomantic results -- i.e., a way of thinking that is not based on the representation of concepts.

Let us now turn to the two types of "cue" recognized in Onomantics: tags and notations. They can be compared with the "terms" and "symbols" recognized in Terminology.

{1.2.1} Tag

A. The Notation. The notation for 'tag' is {1.2.1} and that for 'term' is [5.3.1.2]. This classes each as one of the major forms of "cue" {1.2} and of "designation" [5.3.1]. Its notation puts 'term' at a lower hierarchic level in Terminology than 'tag' in Onomantics. However, both are major items in the taxonomy of succinct forms of concept representation. The differences between them become clear when one looks at their definitions (descriptions).

B. The Description. The definition of [5.3.1.2] term given in ISO 1087 resembles that for "tag" {1.2.1} in Onomantics. However, the inclusion of "linguistic expression" imposes a false limitation since concepts can be represented by non-linguistic symbols as well as by words. Consider, for example, the alphanumeric codes used for chemical concepts, e.g., H₂O, or the symbols used for "plus" (+), "minus" (-), decimal point (.), ampersand (&) and at (@). No doubt most tags are, indeed, linguistic expressions, but sometimes symbols are also used to represent concepts. Consequently, we may say that some, but not all, tags are "terms". Tags include symbols as well as linguistic expressions -- why should "terms" exclude them if they are intended to embrace all the forms used to represent a concept?

A further complication arises because the definition for "terms" [5.3.1.2] includes a distracting stipulation limiting terms to linguistic expressions "in a special language". Since the definition of "symbol" [5.3.1.1] does not mention "special language" one has to wonder if the glossators actually intended to restrict the application of "terms" to special languages while letting "symbols" designate concepts in both special and general languages. In Onomantic usage, concepts can be represented in any kind of language although, obviously, more care is required when using both terms and symbols in special languages.

C. The Tag. Unfortunately, there is another problem with term. As used by lexicographers, this word refers to any word, phrase, or affix that is a lexeme and, accordingly, should head a dictionary entry -- see the discussion following Schedule II. Terms, in this sense, are almost always polysemic as anyone who looks at a general dictionary can easily see: 'term' itself is a polyseme whose dictionary entry reports almost a score of meanings. To use "term" for a designator that represents a specific concept (as specified in [5.3.1.2]) is to stipulate an unusual meaning for the word.

Like the new meaning given to "definition" [4.1], as noted above, this stipulation adds to the number of meanings of a word often used in the same discourse community to mean something else -- this makes it equivocal. If terminological usage were always separate from lexicographic usage, ambiguity might be avoided. However, the two fields overlap at many points and it is difficult, therefore, to keep the semantic meaning of a lexicographic term (lexeme) clearly separate from its terminological use to mean a designator (tag).

{1.2.2} Notation

A. The Notation. As noted above, the location of [5.3.1.1] "symbols" in ISO 1087 shows that the notion is viewed as one of three forms available to designate concepts [5.3.1]. However, the criteria for identifying a "symbol" are formal, not functional: they fail to identify the fundamental distinction made in Onomantics between representing concepts by a short expression (tag) and by a systematizing symbol (notation). The definition for "symbol" [5.3.1.1] identifies it as a non- linguistic form by contrasting it with "term" [5.3.1.2] which is defined as a type of "linguistic expression". It even fails to mention its most important characteristic: locating concepts in a system.

The non-linguistic property of symbols is basic for lexicographers who cannot write entries for symbols (as noted above following Schedule II) but it is irrelevant for terminologists because symbols can be used not only in notations but also in tags. When terminologists select a form to serve a purpose, they need not worry about the type of form to be used. By contrast, when lexicographers choose items to enter in a dictionary, the forms they take determine their treatment. In short, the notations used in ISO l087 reveal that the gloss for symbol [5.3.1.1] expresses a formal distinction that is fundamentally different from the functional characteristics specified in the onomantic gloss for notation {1.2.2}.

B. The Description. These conclusions are supported by the definitions provided in ISO 1087. Symbols are defined, essentially, as non-linguistic forms, like A, B and C, or 1, 2 and 3. No doubt, notations typically take the form of symbols, but many symbols (as noted above) also represent concepts. This means that symbols can be used both as tags and notations.

To clarify this relationship, consider two formal categories: A (terms as linguistic expressions) and B (symbols as non- linguistic items). Now sub-divide symbols into two classes, A1 (cardinal symbols, like 2=two) and A2 (ordinal symbols, like 2=second). Next, think of the contrast between M (tags, as designators of concepts) and N (notations, as pointers to the location of a concept in a system). We can then say that:

M = A + B1 -- tags may be linguistic (terms) or non-linguistic (cardinal symbols)

N = B2 -- notations are always ordinal symbols

Consider that, in their written form, numbers can be read in two ways: "1" may mean "one" or "first"; "2" can mean "two" or "second", etc. One sense of every symbol is cardinal and a second sense is ordinal. The same distinction applies to letters of the alphabet, even though they are pronounced the same way when used cardinally or ordinally. Consider that, in a notation scheme, "A" may represent the first item in a list, whereas the same letter can also be used to represent an article, its shape, a chemical element (angstrom) or it may be an abbreviation for "answer" or "ampere". Similarly, "H" can mean the 8th item in a list, the shape of a beam, or Hydrogen. Only when "H" is used ordinally to mean the eighth item in a set can it appear as a notation (or part of a notation). In its other (cardinal) senses, by contrast, "H" can be a tag (term) for a concept. The same is true of numbers and alphanumeric combinations. In this article, {1.2.1} and [5.3.1.1] are notations written with symbols, but the same symbols can also be used to tag concepts.

From an onomantic point of view, it is both useless and obvious to identify symbols as non-lexical signs. What is essential is that the relations between concepts in a system of concepts should be identified, and that is what notations do by means of ordinal symbols. The fact that both lexemes and cardinal symbols are used as tags is, no doubt, true but unimportant onomantically -- the function they perform when representing concepts is decisive. The distinction made in ISO 1087 between "terms" and "symbols" fails to make this point.

As noted above, after Schedule II, the lexical distinction between words that can be spelled/alphabetized and symbols that cannot is decisive, but this is not an important point in Onomantics and, I think, in Terminology also. It is, of course, a fundamental lexicographic distinction because dictionaries offer entries for terms (linguistic units) but not for symbols.

C. The Tag. All the dictionary senses of "symbol" include the point that symbols represent something but none of them point to the ordering function of notations. Nevertheless, the systematic relations between concepts is often mentioned in Terminology as an essential feature. How should such relationships be indicated? Admittedly, "notation" is a polyseme with several meanings: among documentalists, it refers both to a classification scheme and to code "numbers" in such a scheme. In the usage presented in {1.2.2} each notation designates a concept and only one concept and it also places that concept in the framework of a system of concepts.

This is not to say that terminologists have no interest at all in the various forms that can be used to represent concepts. When choosing between the different forms that might be used to represent a concept, for example, it may be quite important to recognize that some forms will be more acceptable and less ambiguous than others. This consideration leads me to think that a separate category in ISO 1087 ought to be set aside for the identification of various available forms such as words, phrases, affixes, symbols, and graphics, plus the relevant semantic notions such as polysemy, monosemy, homophony and synonymy.

However, as noted in Part I, I think such information ought to be presented separately as a set of borrowed concepts in which definitions taken from existing dictionaries would be used in place of new concept descriptions that, as illustrated in ISO 1087, often stipulate new and confusing meanings for these words. In such a context, it would be relevant to point out that ordinal symbols are non-linguistic forms that can be used to place concepts in a system of concepts. By contrast, of course, lexemes and cardinal symbols are used to represent a concept succinctly (i.e. without describing it) and texts (but not sentences) are used to describe concepts by specifying their essential characteristics.

To conclude, I believe that all of the three types of designation entered in ISO 1087 at [5.3.1.1/3] are essentially incorrect and ought to be replaced by different concepts and tags. Perhaps I should stop here. However, much more can and needs to be said to establish, more broadly, the important differences between Onomantics and Terminology. In Part III of this project, I will discuss them under three main headings: the format of a gloss, the borrowed vocabulary of Terminology; and some missing concepts needed by terminologists. In conclusion, I will speculate about the main reason why Onomantics and Terminology are so different.