Some of the items in the OI! Project lists — see the top items in the survey here — are initially puzzling. For instance, the top item in the IANW (Include All Necessary Words) list — discussed in 12 handbooks — appears in our files as “very + PSP” (past participle), though in my posting I expanded the label to identify the word — much — that is at issue in the handbook discussions. That is, the problem with very surprised, according to some of these discussions, is that it is “missing” the necessary much; the expression should be very much surprised.
I’ll get back to this particular case in a moment. But first, some general comments.
The OI! Project looks at two pieces of “umbrella advice” — Omit Needless Words and Include All Necessary Words — under which a large number of disparate specific proscriptions can be subsumed. All this advice, you’ll note, is about words: using too many (ONW) or too few (IANW). Wherever possible, the usage handbooks couch their descriptions of grammar and usage offenses in terms of word use, rather than choice of construction; they try to blame it on a word. (That allows the handbooks to be organized in dictionary format.)
Of course, not all perceived offenses can be reduced to simple matters of too many or too few words. Some of these offenses have to be treated at a higher, constructional level: split infinitives, dangling modifiers, stranded prepositions, and so on. But insofar as possible, the offenses are blamed on specific words — so the problem with very surprised is the fault of the missing much, rather than a condition on the modification of PSPs.
Blame It On a Word is part of a larger view about the nature of language, the view that “a language is just a big bag of words” (as Geoff Pullum put in on Language Log back in 2003; follow-up here). Geoff’s prime target in these postings was the extraordinarily widespread idea that what’s really interesting about a language is what words it has — and how many it has. While not denying that lexical items are important resources of a language, Geoff was objecting to the idea that they were pretty much the whole business. Instead, Geoff wrote in 2003 (expressing a point of view that he and Barbara Scholz had published about),
Natural languages are much better thought of as systems of conditions on the structure of expressions (words, phrases, sentences).
Yes, the phrases and sentences use words, but as raw materials, a point I emphasized in a Language Log posting on the use of wordsmiths to refer to writers.
[These can be hard things to talk about. Words have a phonetic/orthographic reality for speakers that makes them relatively easy for people to talk about, but more abstract patterns in the structure of expressions aren't at all easy for people to grasp.]
Now back to very surprised, a topic I posted on here back in January, prompted by a column of Jan Freeman’s on usage advice that hasn’t aged well, in this case a proscription from a 1907 book of usage advice by Josephine Turck Baker:
“Don’t say ‘I am very pleased to see you.’ Say ‘I am very much pleased to see you, or I am pleased to see you.’ Note. – Very cannot directly modify a verb, and, hence, not its past participle.
It turns out that there’s a considerable, and very tangled, history of the proscription, as you can see from the excellent discussion in MWDEU (first section of the very entry).
MWDEU‘s story begins with the treatment by Fitzedward Hall in Modern English (1873), who refuted assertions that expressions like very pleased and very delighted were Americanisms by quoting very concerned from 1760 “and other similar 18th- and 19th-century examples, all British.”
Eventually, discussion settled on the question of when past participles are functioning as adjectives (in which case they can be modified by very) or as verbs (in which case they cannot). From MWDEU:
By the time of Viztelly 1906, the question had become one of propriety; making exception for those participles established as adjectives, he found that “it is now thought more grammatical to interpose an adverb between the participle” and very. Similar advice has since appeared in many usage books. The adverb most often recommended was and is much.
Some samples from the OI! files:
M. Alderton Pink (A Dictionary of Correct English, 1948) says that unless it is used as a pure adjective (a very surprised expression), the expression should be very much (p. 199)
V. H. Collins (The Choice of Words: A Book of Synonyms with Explanations, 1952) pronounces that unless a past participle is established as an adjective, very requires much (p. 191)
Maurice H. Weseen (Words Confused and Misued, 3rd ed., 1952) tells us that very interested (or any other past participle) is “not strictly correct” though it’s “in frequent use”; the expression should be very much interested (p. 294)
Michael Swan (Practical English Usage, 1980) writes that with some past participles [Swan doesn't say which], very is not possible and advises using much or very much instead (secs. 453.4, 618.1)
Bryan Garner (Garner’s Modern American Usage, 2003) notes that conservatives say these expressions need the “intervention” of “an insulating much“, adding that many past-participial adjectives have lost their verb-ness, but when you’re unsure you should change very to much or insert much (p. 816)
The history of the usage advice on this point is so tangled because (a) new past participles take on adjectival function over time (some usage authorities change their opinions on specific cases in succeeding editions of their handbooks) and (b) writers have considerable trouble deciding when particular past participles are “established as adjectives” (see MWDEU‘s critique of Fowler’s Dictionary of Modern English Usage (1926)).
The central question in all of this, then, doesn’t have to do with very, or for that matter much, but with the status of particular past participles as potentially adjectival in function.
The fixation with very is a bit puzzling to me, since very is merely the most frequent degree adverbial modifying adjectives, and there are plenty of others. Take extremely, for instance. It can modify adjectives (I was extremely happy at the announcement) but not verbs (*The announcement extremely astonished me). However, many people find it fine with past participles (I was extremely astonished by the announcement). Yet no one seems to have objected to this usage (in the way so many object to very), and certainly no one has suggested inserting much (I was extremely much astonished by the announcement — which strikes me as considerably more dubious than the version without much).