Nothing gives me greater pleasure than to greet you as our guest of honour. There is nothing more refreshing than a good swim.
The parallelism of functions between the two forms of comparison (the comparative degree and the superlative degree) in such and like examples is unquestionable.
As we see, the elative superlative, though it is not the regular superlative in the grammatical sense, is still a kind of a specific, grammatically featured construction. This grammatical specification distinguishes it from common elative constructions which may be generally defined as syntactic combinations of an intensely high estimation. E.g.:
an extremely important amendment; a matter of exceeding urgency; quite an unparalleled beauty; etc.
Thus, from a grammatical point of view, the elative superlative, though semantically it is "elevated", is nothing else but a degraded superlative, and its distinct featuring mark with the analytical superlative degree is the indefinite article: the two forms of the superlative of different functional purposes receive the two different marks (if not quite rigorously separated in actual uses) by the article determination treatment.
It follows from the above that the possibility of the most-combination to be used with the indefinite article cannot in any way be demonstrative of its non-grammatical character, since the functions of the two superlative combinations in question, the elative superlative and the genuine superlative, are different.
Moreover, the use of the indefinite article with the synthetical superlative in the degraded, dative function is not altogether impossible, though somehow such a possibility is bluntly denied by certain grammatical manuals. Cf.: He made a last lame effort to delay the experiment; but Basil was impervious to suggestion.
But there is one more possibility to formally differentiate the direct and dative functions of the synthetical superlative, namely, by using the zero article with the superlative. This latter possibility is noted in some grammar books (Ganshina, Vasilevskaya, 85). Cf.: Suddenly I was seized with a sensation of deepest regret.
However, the general tendency of expressing the superlative dative meaning is by using the analytical form. Incidentally, in the Russian language the tendency of usage is reverse: it is the synthetical form of the Russian superlative that is preferred in rendering the dative function. Cf.: слушали с живейшим интересом; повторялась скучнейшая история; попал в глупейшее положение и т.д.
Let us examine now the combinations of less/least with the basic form of the adjective.
As is well known, the general view of these combinations definitely excludes them from any connection with categorial analytical forms. Strangely enough, this rejectionist view of the "negative degrees of comparison" is even taken to support, not to reject the morphological interpretation of the more/most-combinations.
The corresponding argument in favour of the rejectionist interpretation consists in pointing out the functional parallelism existing between the synthetical degrees of comparison and the more/most-combinations accompanied by their complementary distribution, if not rigorously pronounced (the different choice of the forms by different syllabo-phonetical forms of adjectives). The less/least-combinations, according to this view, are absolutely incompatible with the synthetical degrees of comparison, since they express not only different, but opposite meanings.
Now, it does not require a profound analysis to see that, from the grammatical point of view, the formula "opposite meaning" amounts to ascertaining the categorial equality of the forms compared. Indeed, if two forms express the opposite meanings, then they can only belong to units of the same general order. And we cannot but agree with B. A. Ilyish's thesis that "there seems to be no sufficient reason for treating the two sets of phrases in different ways, saying that 'more difficult' is an analytical form, while 'less difficult' is not" [Ilyish, 60]. True, the cited author takes this fact rather as demonstration that both types of constructions should equally be excluded from the domain of analytical forms, but the problem of the categorial status of the more/most-combinations has been analysed above.
Thus, the less/least-combinations, similar to the more/most-combinations, constitute specific forms of comparison, which may be called forms of "reverse comparison". The two types of forms cannot be syntagmatically combined in one and the same form of the word, which shows the unity of the category of comparison. The whole category includes not three, but five different forms, making up the two series — respectively, direct and reverse. Of these, the reverse series of comparison (the reverse superiority degrees) is of far lesser importance than the direct one, which evidently can be explained by semantic reasons. As a matter of fact, it is more natural to follow the direct model of comparison based on the principle of addition of qualitative quantities than on the reverse model of comparison based on the principle of subtraction of qualitative quantities, since subtraction in general is a far more abstract process of mental activity than addition. And, probably, exactly for the same reason the reverse comparatives and superlatives are rivalled in speech by the corresponding negative syntactic constructions.
Having considered the characteristics of the category of comparison, we can see more clearly the relation to this category of some usually non-comparable evaluative adjectives.
Outside the immediate comparative grammatical change of the adjective stand such evaluative adjectives as contain certain comparative sememic elements in their semantic structures. In particular, as we have mentioned above, here belong adjectives that are themselves grading marks of evaluation. Another group of evaluative non-comparables is formed by adjectives of indefinitely moderated quality, or, tentatively, "moderating qualifiers", such as whitish, tepid, half-ironical, semi-detached, etc. But the most peculiar lexemic group of non-comparables is made up by adjectives expressing the highest degree of a respective quality, which words can tentatively be called "adjectives of extreme quality", or "extreme qualifiers", or simply "extremals".
The inherent superlative semantics of extremals is emphasized by the definite article normally introducing their nounal combinations, exactly similar to the definite article used with regular collocations of the superlative degree. Cf.: The ultimate outcome of the talks was encouraging. The final decision has not yet been made public.
On the other hand, due to the tendency of colloquial speech to contrastive variation, such extreme qualifiers can sometimes be modified by intensifying elements. Thus, "the final decision" becomes "a very final decision"; "the ultimate rejection" turns into "rather an ultimate rejection"; "the crucial role" is made into "quite a crucial role", etc.
As a result of this kind of modification, the highest grade evaluative force of these words is not strengthened, but, on the contrary, weakened; the outwardly extreme qualifiers become degraded extreme qualifiers, even in this status similar to the regular categorial superlatives degraded in their elative use.
1. Ilyish B. “The structure of modern English”, M, 1971
2. Bloch M. “The course in the English grammar”, M, 1983