1. Daniel Jones (1932 §284) described "English Vowel No. 4" as "roughly intermediate between" Cardinal Vowels 3 and 4. Wells (1982:2.2.3) used exactly the same description ("approximately halfway between cardinals 3 and 4") for "mainstream RP". Gimson (1962:7.12) rather curiously preferred to suggest that the norm in his day was distinctly closer ("just below the half-open position"). This decision was possibly arrived at by taking as norm the spatial mean of the spread of articulations within the community of speakers he had in mind rather than the most ordinary-sounding value. I now much regret not quizzing him on this as I did on so many other matters. His own habitual value, which I was able to observe on very many occasions, didn't strike me as particularly close. I don't recall it as materially different from for example David Abercrombie's ash. Abercrombie's judgement of his own usual value (personal communication), with which I could very readily concur, was that it was less close than half way between open and open-mid and somewhat less than fully front. Fortunately there are a fair number of recordings of both of them still in existence for anyone who should want to check these impressions.
2. Many older people in particular could be heard, as BBC archive recordings from the forties and fifties (and sixties in some cases) can remind us, employing values as close as Cardinal 3. Gimson (1962:7.12) labelled this type as "refined". It was not just socially conspicuous but also beginning to sound rather old-fashioned at the time he was writing and by the end of the seventies had become markedly so. British films made even as late as the fifties now often strike many people as quite droll when they, as so often, abound in characters displaying the quaintly "refained" qualities of this vowel.
3. There are indications that a very close ash more like [ɛ] than [æ] was felt to be affectedly posh from various literary sources. In his quasi-autobiographical novel Kipps published in 1905 the Kent originary H. G. Wells represents his eponymous chief character as using the expletive Dash! (eg p.227) initially but, when attempting a more socially elevated style, repeatedly converting it to Desh! (eg pp 229, 231, 252) and using expressions like enegram (p.266), treshy novels and Fency! (p.299), and merried man (p.346).
4. Gimson (1962:7.12) also included a parenthetic remark about the use of "a more relaxed /æ /... in the region of Cardinal [a] ... heard amongst children in the south of England who ... later in life adopt the tenser and closer variety of /æ/". In 1970 he added a further comment that "Such a lowered /æ/ is maintained by many young women ... The result can be a confusion of /æ/ and /ᴧ/". He was certainly right that one particularly noticed the lowered and retracted varieties among fashionable young women. My personal label for it at the time was "deb's ash". A classic examplar of it was the actress Anna Massey who in 1955 at the age of eighteen created the title rôle in the stageplay The Reluctant Debutante. By the end of the seventies it was becoming unremarkable. Wells (1982:4.1.6) quoted the popular novelist Jilly Cooper's references to "Flat A" in a Sunday Times article of 1978. Kingsley Amis, who of twentieth-century British writers showed an outstanding interest in representing speech characteristics, in Girl 20, a novel of 1971, obviously had the usage in mind when he wrote 'What a terribly nice fluht' she said using the then fashionable throaty vowel. Princess Anne, who was born in 1950 and educated at a girl's public school, exhibited quite a retracted variety.
5. The tendency to retract this vowel has in general not progressed much further than towards half way between front and central but exceptionally it has for some become a centralised back vowel, no doubt in some cases differentiated from such speakers' /ɑː/ only by length. An example of a public figure who has had an extreme such value is Baroness Emma Nicholson (of Winterbourne born 1941) who has a markedly upper-class background and was educated at the very fashionable girls' public school St Mary's School, Wantage. Her extreme value for this vowel may have had some connection with hearing impairment with which she is known to have suffered.
6. To return to Gimson's comment "The result can be a confusion of /æ/ and /ᴧ/", one has to agree that for the listener it can theoretically at least produce situations where one may be uncertain of the speaker's intention. Curry and Carrie might for example possibly constitute such a pair. I can readily attest that I have often noted occasions when I should have been in doubt, without the almost invariably available ample contextual clues, whether a speaker extremely well known to me intended one or the other phoneme. However, I have never noticed any speaker produce a rhyme when uttering expressions like catgut, gun-ban, rucksack, suntan, wax ducks etc and one wonders whether Gimson's perhaps ambiguous remark was meant to imply such a possibility. Wells (1982:ibid), while not exactly contradicting Gimson's comment, goes no further than to say that it "may even be the case that for some speakers" the vowels in question are merged.
7. Cruttenden's admirable 1994 re-editing of Gimson, also not actually contradicting the original account, similarly dealt with this question very cautiously. He referred to the "around" Cardinal [a] version of ash current among "many younger speakers" and added "Since the vowel /ʌ/ has had a tendency in recent years to move forward towards Cardinal [a], this may occasionally result in a neutralization of /æ/ and /ᴧ/. More often, however, the lowering of /æ/ results in a retreat of /ᴧ/ towards the central region" (§8.9.4 p.103). This retreat of /ᴧ/ is exactly the way I have been inclined for some time to describe the phenomenon.
8. Wells (1982:4.1.5 and 4.3.7) has a good deal to say about the common lengthening of ash in a good number of words and the fact that some few speakers seem to have an /æ ~ æ:/ phonemic opposition in their speech operating in a small number of words. What I find of particular interest is the occurrence of group of words with, for some speakers, long values of ash in a pre-enclitic syllable. Such syllables most often resist the lengthening that very widely occurs in monosyllables. Thus GB speakers frequently lengthen the ash in bad, can, hand, man etc but not necessarily in badly, candid, handy, manner etc. There are, however, quite a number of speakers who exhibit such pre-enclitic lengthenings who nevertheless have accents which, observantly and carefully listened to, reveal no (other) non-general characteristics. What is more, these lengthenings in such speakers apparently do not strike others as regionalisms, but presumably only as personal speech features. It is perfectly true that the same types of lengthenings may occur in the speech of people with some degree of obvious regional affiliation. When they do so, that regional affiliation is regularly southern and usually southeastern as far as I have been able to observe.
References
Amis, Kingsley (1971) Girl 20. London: Jonathan Cape Ltd
Cruttenden, Alan (1994) Gimson's Pronunciation of English. Fifth Edition. London: Edward Arnold.
Gimson, A. C. (1962, 1970) An Introduction to the Pronunciation of English. London: Edward Arnold.
Jones, Daniel (1932) An Outline of English Phonetics. Leipzig:Teubner.
Wells, H. (1905) Kipps. London: Penguin Books.
Wells, J. C. (1982). Accents of English. (3 vols). Cambridge, UK: CUP.