Introduction
If you look through a modern guide to English usage, you will probably find that it has something to say about phrasal verbs. It might be a warning not to use certain phrasal verbs in certain contexts. For example, Allen (Reference Allen2005: 181–90) offers a table of ‘more formal alternatives to those phrasal verbs that can sometimes be too informal for writing’, with suggestions for replacing sum up with conclude and step down with resign, among others. In many cases, it will be a warning about phrasal verbs where the adverbial particle adds little semantic content to the verb: The Chicago Manual of Style advises writers to ‘avoid the phrasal verb if the verb alone conveys essentially the same meaning – e.g. rest up is equivalent to rest’ (2003: 174). It might attribute such usages to American English, as in Evans' (2000: 54–5) comment that phrasal verbs such as win out, stop off and check up on, which ‘grow like toadstools’, are ‘American parasites’. It might be a positive comment, such as Bryson's note that phrasal verbs are ‘one of the most versatile features of English’, but if so, it will probably be qualified: Bryson adds that in many cases the added particles ‘are merely a sign of careless writing’ (Bryson, Reference Bryson2002: 156–7). What is it about phrasal verbs that provokes such comments? By examining grammars, usage books, dictionaries and other materials since the eighteenth century, I will discuss changing attitudes towards phrasal verbs and how they fit into the context of broader opinions about language.
Common and uneducated?
George Campbell, author of the highly influential The Philosophy of Rhetoric, compared English (and other modern languages) with Latin and Greek as follows:
Our modern languages may […] be compared to the art of carpentry in its rudest state, when the union of the materials employed by the artisan, could be effected only by the help of those external and coarse implements, pins, nails and cramps. The ancient languages resemble the same art in its most improved state, after the invention of dovetail joints, grooves, and mortices, when thus all the principal junctions are effected by forming properly the extremities or terminations of the pieces to be joined. (Campbell, Reference Campbell1776: Vol. II, 413)
Campbell was criticizing English as an analytic language: its grammar is largely based on word order and separate words such as prepositions (‘pins, nails and cramps’). Latin and Greek, on the other hand, are synthetic languages which use verb and noun endings (‘dovetail joints’) to indicate tense, case and other relationships. Campbell was voicing the opinion of his time: Latin in the eighteenth century was the language of learning and of elevated prose, and was thus seen as inherently superior.
This elevation of Latin meant that phrasal verbs – which are usually composed of monosyllabic verbs and particles of Germanic origins – have often been seen as the poor relation of their Latinate equivalents, prefixed verbs. One nineteenth-century grammar complained that ‘there is still a great deficiency of single words, especially in the verbs, as appears in the custom of associating prepositions by annexation rather than incorporating them by composition: thus we use such forms as “to think of”, “to try on”, “to come by”, “to laugh at”, “to put off”, “to bring under”, and numerous others’ (Parminter, Reference Parminter1856: 153). The single-word alternatives to these phrasal and prepositional words are Latinate: put off could be replaced by postpone; laugh at by deride, and so on.
Unfavourable comparisons of phrasal verbs with their Latinate equivalents continue into the twentieth century. In The Modern English Verb-Adverb Combination (1920), the first book devoted solely to phrasal verbs, A. G. Kennedy takes a surprisingly low view of his subject: ‘[t]he development of these combinations is essentially a process of the common, relatively uneducated, mind’ (Reference Kennedy1920: 40), and ‘much of the usage [of phrasal verbs] is a result of linguistic laziness’ (1920: 44). In a later article, he deplores the ‘growing avoidance of many special verbs such as recover “to get over,” exhaust “to use up,” examine “to look over,” […] this disuse of such verbs threatens to cut down the active vocabulary of English very materially during the next few generations’ (Reference Kennedy1933: 6). Although Kennedy does not specify the reasons for his preference, it is evident that the ‘special’ verbs (recover, exhaust, examine) are Latinate. (He may also be referring to the fact that the Latinate verbs have single, special meanings, while the ‘lazy’ phrasal verbs have many meanings; I will return to this point below.) However, the fate of native English words improved throughout the twentieth century. In part this was due to campaigns for ‘pure’ or ‘plain’ English, most famously voiced by George Orwell, who argued against ‘pretentious, Latinized style’: ‘Bad writers, and especially scientific, political and sociological writers, are nearly always haunted by the notion that Latin and Greek words are grander than Saxon ones’ (Orwell, Reference Orwell1946). Choosing ‘Saxon’ over Latinate terms often means choosing a phrasal verb over a Latinate equivalent, in direct contrast with the alternatives proposed by Kennedy, above. For example, according to the Plain English Campaign's website, it is better to avoid calculate, comprise, discharge and exclude, and to choose work out, make up, carry out and leave out instead. This is also linked to the decline in classical learning; Latinate verbs are no longer transparent, and can seem lifeless. One early-twentieth-century commentator on American usage mentioned the American preference for phrasal verbs such as turn down rather than Latinate verbs such as reject, and points out that ‘[t]he advantage of the American idiom [i.e. phrasal verbs] is that it preserves the vividness of the metaphor, while in the English idiom [i.e. Latinate verbs] one has almost come to forget that the term employed uses any metaphor at all’ (Horwill, Reference Horwill1936: 195). That is, for many people the metaphors in Latinate verbs are dead: reject is no longer analysed as ‘throw back’, but is simply an opaque term.
Also, in contrast with Kennedy's ascribing phrasal verb usage to a particular set of users (uneducated, lazy ones), usage guides nowadays tend to compare phrasal verbs and Latinate verbs in terms of register. Some state baldly that ‘Phrasal verbs, which are verbs followed by prepositions, are too informal for academic writing’ (Hong Kong Polytechnic University). Other guides are more subtle in their recognition that some phrasal verbs, such as set out, carry out, and sum up, are acceptable in more formal registers, while others, such as put off and bring up, are not. However, even this indicates a certain negativity which is still attached to phrasal verbs. After all, the same could be said of most grammatical classes. Some adjectives are suitable for formal writing (overweight, unusual, wealthy) and others are not (podgy, wacky, minted), but this is not generally mentioned in sections on adjectives in usage guides. Some usage books – especially those written for non-native speakers of English – take a more positive approach. Since learners of English often find phrasal verbs difficult, and overuse Latinate equivalents even in informal situations (especially if their mother tongue is a Romance language), usage books for English learners often advise learners to choose phrasal verbs as the most natural option. For example, at the entry for apply meaning ‘put something onto a surface’, Cobuild English Usage writes that ‘In conversation and in most kinds of writing, you do not say that you “apply” something. You say that you put it on, rub it on, rub it in or spread it on’ (Sinclair, Reference Sinclair2004). This is a more positive approach to the register of phrasal verbs: rather than being discouraged in formal situations, they are encouraged in informal ones. It is debatable whether usage books for non-native speakers have an influence on those written for native speakers – they are quite different in content and focus – but the increased awareness of the uses of phrasal verbs in contrast to Latinate ones in an EFL context may be having a positive effect on their reception generally.
Allowing anything to mean anything
One of the most striking features of phrasal verbs is their tendency to have many meanings. Indeed, the book which introduced the term polysemy to English used phrasal verbs as an example:
We now see the reason why polysemy is found so often in small words to an extent which would not be tolerated in longer words. This is particularly frequent with short verbs, […which] are frequently used in connexion with adverbs or prepositions in such a way that the meaning of the combination can in no wise be deduced from the meaning of each word separately, cf. for instance put in, put off, put up (put up with), make out, make for, make up, set down, set in, set out, set on, set up (this with some forty subdivisions), give in, give out, give up, &c. (Jespersen, Reference Jespersen1928: 26)
This feature does not always invite praise. Polysemy was often criticized in the eighteenth and nineteenth centuries: one usage guide asserted that ‘when we find in a language one word serving many needs, we may be sure that that language is the mental furniture of an intellectually rude and poverty-stricken people’ (White, Reference White1870: 81). Following this, we find critical comments about polysemous phrasal verbs. For example, Partridge (Reference Partridge1947: 326) complains about a reference to a bus taking off, arguing that ‘“take off” has several meanings already, both active and passive, and, unless we are to allow anything to mean anything and everything to mean everything else, some precision is advisable’.
Furthermore, we find many comments about the illogical nature of phrasal verbs which run counter to the ‘true’ meaning of either the verb or the particle. For example, one nineteenth-century usage guide goes into great detail about the illogicality of up in ‘taken up for a crime’ and lock up:
It is only a chance that his prison chamber may be up, instead of down; that is, elevated above the street, on which he was taken, instead of being sunk below it. If his crime be murder, or any other of a heinous stamp, he must be content with the lowest dungeon the jail can afford; and when settled in it, he must certainly be taken down, and not up. Fancy him brought to a lock-up house, and secured in the strong barred room on the first floor, while his guard sits over his head, on the second floor. One of them would easily say, ‘he is locked up at last.’ ‘Where is he locked up?’ - ‘Downstairs.’ (Anon, Reference Anon1826: 90–1)
Such comments relate to the etymological fallacy: the belief that every word has only one true meaning. That is, it is assumed that up must mean ‘to a higher level’, whereas it has many other meanings: included in the 33 numbered senses in the Oxford English Dictionary (OED) (with many further subdivisions) are ‘To the notice or consideration of a person or body of persons (spec. of one in authority)’ (as in taken up for a crime) and ‘Into a close or enclosed state’ (as in lock up).
This fallacy was at its height in the nineteenth century, when writers like John Horne Tooke and Charles Richardson attempted to determine one single meaning for every word in English. Nowadays few people claim that there is anything inherently wrong with a word that has more than one meaning, although it is still common to find despairing remarks about the polysemy of such phrasal verbs as run down, or the apparent lack of logic in the fact that you can either fill out or fill in a form, with no difference in meaning.
Obesity in the language
It is often stated that Samuel Johnson was very critical of phrasal verbs: this myth seems to have originated with Logan Pearsall Smith who, in the tract in which he coined the term phrasal verb, claimed that because Johnson wanted ‘to do away with “grammatical irregularities”’, he ‘naturally disapproved of these idiomatic combinations’ (Smith, Reference Smith1923: 58). In fact, in the whole of Johnson's A Dictionary of the English Language (1755), there are only sixteen phrasal verbs to which Johnson attaches a negative label – not many in a dictionary which includes hundreds of phrasal verb senses. However, the one type of phrasal verb which Johnson does criticize is the redundant phrasal verb. At fill up he notes that ‘Up is often used without much addition to the force of the verb’; at fall down he remarks that ‘down is sometimes added to fall, though it adds little to the signification’; he also comments on the redundancy of beat up, copy out, fell down and sue out. Other eighteenth- and nineteenth-century texts make similar comments about particular phrasal verbs: Campbell (Reference Campbell1776: Vol II, 278) comments of return back and come forth that back and forth ‘serve neither for ornament, nor for use, and are therefore to be regarded as encumbrances’; Noah Webster (Reference Webster1828) criticizes the redundancy of phrasal verbs such as fill up, breed up, bloat up, pen up and pucker up; and even the first edition of the OED, intended as a beacon of objectivity, remarks that ascend is ‘occas. emphasized by a redundant up’ and enter is found ‘with pleonastic in’.
In the nineteenth century there is an awareness of particular particles – especially up – being used redundantly. Of open up, one usage book asks:
Can any English scholar inform anybody else what is the propriety of ‘up’ in those and in a thousand similar instances? No doubt, ‘up’ is a little word, and it may often be overlooked in a crowd; but it has a very ambitious strut, when thus paraded on stilts. (Gould, Reference Gould1880: 108–9)
By the mid-twentieth century the discussion of redundant phrasal verbs was a staple of usage manuals, and we find such comments as the following.
The trouble is that the baser sort of English-speaker […] thinks that it is right and clever to add ‘up’ or ‘out’ to any short verb, though the sense is neither enriched nor altered. (Herbert, Reference Herbert1935: 153)
Such comments are based on the assumption that words must have identifiable separate meanings, whereas it is possible that they have different functions. Up, in particular, often has an aspectual (or, more accurately, an Aktionsart or ‘lexical aspect’) function: it shows that the writer or speaker sees the action as complete. For example, if you face a dilemma, you have a dilemma that needs to be dealt with (but you may avoid dealing with it); if you face up to a dilemma, you actually deal with it. If you eat a banana, you consume at least part of it (but you may leave it if you decide half-way through that you are not hungry); if you eat up a banana, you finish it (it would be odd to say ‘I started eating up my banana and then left it’). The same applies to other adverbial particles. You can start to return to something (for example, ‘The refugees started to return to their homes’), but you are unlikely to start to return back, because returning back is a completed act.
American parasites
Before the twentieth century, there were plenty of negative comments about redundant phrasal verbs, but no blame was apportioned to any particular set of users (except perhaps the ‘uneducated’ ones, and the occasional comment that certain combinations were Scotticisms). In the early twentieth century, though, these phrasal verbs began to be associated with American English. A series of letters in The Times (8–16 February, 1933) begins with one reader criticizing the Chancellor's use of the phrase try out in ‘try out the possibilities of these new methods’: the reader ‘cannot believe that [the editor] would put an Americanism into his mouth’ (8 February, 1933, 13). Other readers agree, criticizing the allegedly American cancel out, speed up, slow up, sign up, stand up for and check up, described as ‘American gate-crashings’. Later letters and articles in The Times display a similarly negative attitude: one reader asks: ‘Why does Sir John R. Marriott inflict on us the horrid Americanism of “face up to”? Why can he not simply and tersely “face the facts”?’ (17 August, 1934, 8); another asks, of man up, meet up with and study up on, ‘Must the strong, simple transitive verb, which is one of the main glories of our tongue, become as obsolete in England as it appears to be in America?’ (6 February, 1947, 5). An editorial piece (17 June, 1964, 13) complains about win out, help out and fire out, where
a nasty verbal convention takes the place of a plain word…“Out”, of course, used as an unnecessary auxiliary [sic] comes into this country from the United States, and, while we have been benefited much linguistically and otherwise from the flow of imports from that country, this is one immigrant we can dispense with.
Usage books follow suit: Gowers (Reference Gowers1948: 41–2) writes that redundant phrasal verbs are an ‘infection which […] is spreading across the Atlantic [and] calls for watchfulness’. This belief that redundant phrasal verbs are a particular feature of American English (but are ‘invading’, ‘infecting’ or ‘gate-crashing’ British English) prevails today: as noted in the introduction, Evans (Reference Evans2000: 54–5) uses a similar metaphor when he describes phrasal verbs such as win out, stop off and check up on as ‘American parasites’.
How accurate are these comments? We can look at evidence about particular phrasal verbs in the OED, and find that some of these were indeed first recorded in American English, including check up (on), meet up (with) and win out. Others were in the language long before Americans could be blamed for them: stand up for is first recorded in 1608, in King Lear; help out in 1600. Others are relatively new, but with no evidence of being particularly American in origin or use: all the quotations for man up in the OED are from twentieth-century British texts, while speed up is first recorded in 1894 in the Westminster Gazette, with later quotations from a mixture of British and American texts.
There are plenty of other phrasal verbs which are at least as ‘redundant’, and which are firmly British in origin. Eat up is first recorded in 1535 in a translation of the Bible; gobble up in 1601. The first quotation for cool down in the OED is from a nineteenth-century British book; warm up in the literal sense (warm up food) is first recorded in 1848 in Dickens' Dombey and Son; in the figurative sense (warm up to a subject), the first quotation is from the nineteenth-century British Prime Minister, Benjamin Disraeli. As I have argued above, though, these are not really redundant, and neither are the phrasal verbs that are disparaged as Americanisms. The point of interest is that because people object to ‘redundant’ phrasal verbs, they associate them with American English, and neglect to mention the many British equivalents. It is notable that these objections first arose in the 1930s, around the time that American influence on British culture (especially through music and film) was beginning to be felt and sometimes resented.
Conclusion
Perhaps because they are so flexible – they often have many meanings and can be used in a variety of grammatical patterns, contexts and registers – over the centuries, phrasal verbs have elicited strong opinions in writers of grammars and usage guides. These opinions can be understood by examining the history of ideas about English, including the relationship between English and Latin, attitudes towards meanings, and attitudes towards varieties of English. By reflecting on the history of such linguistic debates, perhaps we can move towards a more balanced view of the much-maligned phrasal verb, through understanding of the history of the varied sources of such debates, and the evaluation of their influence on prescriptivist attitudes.
KATE WILD completed her PhD on English phrasal verbs at the University of Glasgow in 2010. She also worked for five years as an editorial assistant on the Historical Thesaurus of the Oxford English Dictionary and, before that, as an EFL teacher in Poland and Vietnam. Her main research interests are dictionary history and lexical history, and she has published several articles in these fields. Kate currently works as a freelance lexicographer and corpus linguist. Email: katewild2003@gmail.com