‘How “English” are the phrasal verbs really?’ and how did this highly familiar but often ill-described construction evolve ‘from its early history up to the present’? (p. 1). Thim's book, which may be considered less as an ambitious empirical investigation of the topic than as a competent critical survey of its literature, aims at answering these two questions. That these questions are intertwined is clear from the answers they receive in this work.
First, the phrasal verb is not very special to English. While the term phrasal verb ‘is rarely ever used except with respect to English’ (p. 2), Thim points out, as many scholars have done before him, that there are close parallels to be found in other languages, most notably in the other present-day Germanic languages. For instance, the German prefix verb aufgeben ‘give up’ may occur as two separate words, with the prefix auf (a cognate of up) even obligatorily split from the verb by a Direct Object noun phrase (e.g. Sie gab ihre Arbeit auf ‘She gave up her job’). Thim further cites examples of particle verbs from Danish, Dutch, Norwegian Nynorsk and Swedish, as well as from Afrikaans, Faroese, Icelandic and Yiddish, to support the claim that the English verb–particle construction is certainly not an isolated language-specific phenomenon.
Second, these close parallels then help Thim answer the question about the evolution of the ‘English’ phrasal verb, whose origin can be traced back at least to Proto-Germanic. In Indo-European languages more generally, Thim reminds us, prefix verbs and particle verbs have developed out of adverb–verb sequences. And in fact, in various languages belonging to genetically more distant families, particle-like verb prefixes, or ‘preverbs’, have formed out of adverbs, typically derived themselves from previously independent relational nouns. English may look special in that particles now standardly follow the verb, leaving aside exceptional structures of the type In came a strange figure (more on which below). However, in Old English, and still in early Middle English, particles could appear in both preverbal and postverbal position, depending to some extent on clause type (subordinate or main) and finiteness, rather like the separable prefixes we encounter in Continental West-Germanic languages such as present-day Dutch or German.
Thim strongly rejects the widespread misconception ‘that there must have been a “rise” of the phrasal verb’ (p. 145), that is, that the phrasal verb was a Middle English or later innovation which superseded an earlier pattern, one with inseparable prefixes. Rather, Thim argues, the apparent emergence of postverbal particles in English is nothing but an epiphenomenon of independent, well-established changes in the language system, the most important of which is the long-term shift from basic O(bject)V(erb) order in Proto-Germanic to VO order in Modern English. In Old English, the oldest order, Object – particle – Verb, was typically found in subordinate clauses and occasionally still in main clauses. When the verb started to occupy the second position in clauses (‘V2 movement’), unstressed preverbs that had fused with the verb stem moved along with it as inseparable prefixes. By contrast, preverbs carrying stress (‘particles’) remained independent and stayed behind in final position when the verb itself moved leftward. It was first the finite verb (v) which moved (O prt V v → v O prt V), causing a clausal brace with the non-finite main verb (v . . . V), but the latter subsequently rejoined the finite verb in a process known as exbraciation (v O prt V → v V O prt). Extraposition of especially heavy objects may have played a role in dissolving the brace and further yielded the alternative structure with the particle immediately following the verb (v V prt O).
It is not immediately apparent to the reader to what extent this account offers new insights compared to previous analyses. Some more explicit comparison with for instance Elenbaas's (Reference Elenbaas2007) analysis would have been welcome. On the one hand, Thim discusses the general development of English word order in rather bewildering detail – for example, he provides early fifth-century examples from the Northwest Germanic runic corpus and quantitative data from Denison's (1987) study of early West Saxon clausal patterns. On the other hand, he glosses rather too swiftly over a key diagram, on the causes of changes in particle position in Medieval English (figure 3–2), leaving the reader wondering how to account for a couple of patterns shown there (prt V v O and v prt V O), especially since in the subsequent pages (pp. 105–15) none of the examples which could possibly serve as illustrations for these patterns contains a finite verb form that is distinct from the main verb. Nonetheless, the general idea is clear: particles didn't move from a preverbal to a postverbal position; it is the verb that moved to a pre-particle position.
Thim's book appeared in the same year as Los et al.'s (Reference Los, Blom, Booij, Elenbaas and van Kemenade2012) Morphosyntactic change: A comparative study of particles and prefixes (Cambridge University Press). The two monographs cover much the same ground and I therefore initially intended to compare their claims and analyses in a double review. However, this would have made for an even lengthier piece than this review already is. Besides, Los et al. (Reference Los, Blom, Booij, Elenbaas and van Kemenade2012) has meanwhile been reviewed by Den Dikken (Reference Den Dikken2013) and Hoppermann (Reference Hoppermann2013). At a superficial level of comparison, there is no doubt that Thim's publication is extremely well researched; its list of references totals 38 pages, almost thrice the number of pages with references in Los et al.'s book, which in turn already puts to shame the two pages containing a mere 16 titles treating phrasal verbs and the odd footnote citing an additional publication in Bolinger's (Reference Bolinger1971) influential monograph. Despite this, Thim's book is much less concerned with contemporary linguistic theory than Los et al.’s. It mentions Construction Grammar as a suitable theoretical approach but does little more than pay lip service to it, and though it mentions relevant publications on grammaticalization and lexicalization, it does not actively engage in current debates surrounding these general topics.
Thim's real concerns lie elsewhere. He is interested not just in the development of the construction itself, but also in the origins of certain stubborn, questionable views embraced by linguists, language commentators and lexicographers. He critically re-examines the classic study by Kennedy (Reference Kennedy1920), which he blames for ‘an undercurrent of tacit assumptions about phrasal verbs in most subsequent 20th century research’ that have never seriously been put to the test and which on closer scrutiny turn out to be ‘rather dubious’ (p. 118). Likewise, Thim evaluates the treatment of phrasal verbs in some major syntactic handbooks, which he finds rather confused, and in a number of widely used nineteenth- and twentieth-century historical dictionaries, all of which he considers unsatisfactory: ‘they fail to document the Old and Middle English particle verbs sufficiently and transparently, and thus contribute to the wrong impression that phrasal verbs are an essentially new structure which fully emerges only towards the Modern English period’ (p. 144).
Thim has also conducted interesting philological research into attitudes towards phrasal verbs. He insightfully challenges the received opinion that the construction has always been felt to be ‘native’ or ‘colloquial’. He is at his best when he reveals how textual evidence fails to substantiate the existence of negative attitudes commonly ascribed to the dictionary-maker Samuel Johnson and to style-conscious writers such as the publisher of James Cook's journal and John Dryden, author of the Essay of Dramatick Poesie. Thim concludes that phrasal verbs weren't stigmatized in the late seventeenth and eighteenth centuries or systematically replaced by Latinate alternatives. Unfortunately, as his book is the published version of his 2009 PhD thesis, it apparently has not had the benefit of including findings from Wild's (Reference Wild2010) PhD thesis on perceptions of the phrasal verb in Late Modern English, thus rendering void his claim that ‘the prescriptive attitudes of the 19th century are as yet fairly unchartered [sic] ground’ (p. 233).
The present review deviates from the standard format in that it does not summarize each book chapter in turn. For a useful chapter-by-chapter summary, I refer the reader to Brinton's (Reference Brinton2013) excellent (and generally positive) review in Language. In what follows, I will selectively focus on some further common views which Thim contests and which people studying the construction should do well to stop entertaining. In addition, in the spirit of Thim himself, I will touch upon a few of his own claims which I do not consider entirely correct and which run the risk of being perpetuated too, if left unrefuted. After all, the positions taken in this new book on phrasal verbs deserve the same kind of careful scrutiny that its author applies to claims made in the preceding literature.
1. Phrasal verbs are by definition idiomatic
There is a flourishing market in English language learning resources which make phrasal verbs out to be hard to acquire because of their unpredictable meanings. However, Thim states quite rightly that their meanings ‘range on a cline from purely compositional to highly idiomatic’ (p. 11). For practical reasons, he in fact distinguishes between ‘three, albeit somewhat idealised, semantic types of phrasal verbs, which have been characterized as “literal”, “aspectual” and “non-compositional” (or “idiomatic”)’ (p. 13). The former two types together form a subclass of ‘compositional’ combinations, with either a directional or an aspectual particle. The reason for including these semantically transparent combinations is that ‘only compositional combinations show the full range of syntactic properties typical of phrasal verbs’ (p. 15).
One could of course object that a literal or aspectual particle can be found in non-compositional combinations, too (cf. Jackendoff Reference Jackendoff, Dehé, Jackendoff, McIntyre and Urban2002; Lohse, Hawkins and Wasow Reference Lohse, Hawkins and Wasow2004; McIntyre Reference McIntyre, Dehé, Jackendoff, McIntyre and Urban2002). For instance, in pick something up, the particle is directional but the verb doesn't mean ‘choose’; and in beaver away, the particle has the aspectual value also seen in chat away or type away while the ‘verb’ derives its interpretation of ‘work hard’ partly from its use in a ‘constructional idiom’, a notion which Thim recognizes as useful (p. 58), but then chooses not to adopt (p. 72). A minor objection of descriptive inadequacy can be made with respect to Thim's statement that ‘fronting of the particle is not possible with the aspectualizing particles’ (p. 17; cf. also p. 24: ‘fronting of the particle is only passible in compositional combinations with a directional particle’). In fact, as the following examples from GloWbE (Davies Reference Davies2013) show, continuative aspectual particles (see (1a–b)) do allow this option, and so do certain particles that one can could call ‘literal’ (semantically independent of the verb) but aren't directional (see (1c) (Cappelle Reference Cappelle, Dehé, Jackendoff, McIntyre and Urban2002):
- (1)
(a) I was told in Penang you must eat eat eat! So away I chomped! (GloWbE)
(b) Yet on they dream, the optimists, and on they build. (GloWbE)
(c) Out went the lights; the fleet became dim, dark bulks against an intense blue sky that still retained an occasional star. (GloWbE)
According to Thim, the tripartite semantic categorization of phrasal verbs in present-day English mentioned above can be represented ‘as the result of two historical processes: first, the metonymical evolution of aspectualizing meaning observable in the particles, and second, the further lexicalization of individual complex words’ (p. 182):
Semantic development of verb-particle constructions (= fig. 5–1 in Thim's book)
Thim's proposal is neat in its simplicity, but somewhat stipulative. It is to be doubted that all verb–particle combinations will over time become non-compositional. For instance, for as long as walk in remains recognizable as a verb–particle combination, it may well retain a purely compositional meaning. Further, Thim could have elaborated a bit more on the reason why the shift from a directional to an aspectual meaning has to be understood as metonymical and not metaphorical in nature (Traugott Reference Traugott1991), while a combination as a whole can acquire a (conventionalized) metaphorical use, as will be shown next.
2. Old English particle verbs were semantically compositional
On the basis of the above explanation, it could be assumed that verb–particle combinations in Old English were all still compositional. Even though ‘the vast majority of particle tokens in Old English will have been spatial’ (p. 181), it is not the case that they all were, as is clear from Thim's introductory chapter, where he nicely illustrates the Old English particle verb forþfēran not only with two possible orderings (forð ferde and ferde forþ) but also for each ordering in its literal sense of ‘travel/move away or by’ and its metaphorically derived sense of ‘die’ (cf. also pass away in present-day English). So, the processes shown in figure 1 above aren't locked in time but can be observed to be in force throughout history. Thim argues in various places that particles and the earlier native (Germanic) prefixes, ‘with their similar spatial etymologies, were bound to take similar paths of semantic development’ (p. 195). The processes thus acquire the status of universally valid laws of development, operating across boundaries between languages and across historical periods.
3. Phrasal verbs are word formations
A recurrent debate in the literature on phrasal verbs is whether they are words or phrases. Thim seems to circumvent the issue by calling them, following Booij's (Reference Booij, Dehé, Jackendoff, McIntyre and Urban2002) treatment of separable complex verbs in Dutch, instances of ‘periphrastic word formation’ (p. 62), that is, word forms consisting of more than one word. He then states more precisely that they can ‘be analysed as compounds (compositional phrasal verbs: combinations of a verb and a spatial particle), but also as derivations in those cases where the meaning of the particle is different from the meaning of the free adverb (in particular in aspectual combinations) or has a clearly derivational function, as in [furnitured out, terraced out, sexed up, big up]’ (p. 65).
I would be the last to object to the view that phrasal verbs can be treated by language users as lexically stored items, even with word status (see Cappelle, Shtyrov and Pulvermüller Reference Cappelle, Pulvermüller and Shtyrov2010 for neurolinguistic evidence). Yet we should probably also accommodate their realization as syntactic phrases, since the verb and the particle can notoriously be separated from each other by phrasal material (e.g. let the dogs out, pull the car over), which would be hard to explain if they were always word parts. A solution may be sought in an ‘allostructional’ model, which Thim mentions himself (p. 71), as a way of linking the two word-order patterns of transitive phrasal verbs (V prt NP and V NP prt) (cf. Cappelle Reference Cappelle2006; Goldberg to appear). These two structures could be analysed as two variant realizations of a single lexeme whose morphological or syntactic status is underspecified (Cappelle Reference Cappelle, Bergs and Diewald2009; Farrell Reference Farrell2004).
The category-changing and valency-changing properties of phrasal verbs (for the latter, see especially McIntyre 2007) needn't be taken as support for a derivational word-formation analysis, as Thim does. Syntactic structures, too, can force a non-verb into the verb position (e.g. {debt/junk-food/X-boxing} themselves to death (all of these attested in GloWbE)), or provide arguments not selected by the verb itself (e.g. {laugh/worry} oneself silly, cry oneself to sleep). As it would be implausible to treat silly, into a stupor, to {boredom/death/exhaustion}, etc. as affixes or even as affixoids, I see no reason why one should suggest that particles in similar structures are ‘more or less affixal in nature’, as Thim (p. 62) does (thereby quoting Bolinger Reference Bolinger1971: 112).
4. Particles necessarily carry stress
A recurrent claim in the literature, one which can also be found in Thim's book, is that ‘in phrasal verbs, the particle carries stress’ (p. 68). While this is true if the particle follows the verb directly (thus allowing a distinction to be made between phrasal verbs and prepositional verbs, e.g., Which number did she look úp? vs Which book did she lóok at? (p. 27)), it very often is not the case when the verb and the particle are separated by a non-pronominal object NP, as is shown by these examples from Bolinger (Reference Bolinger1971: 51):
- (2)
(a) Let's take our friénds over.
(b) They shot the whole pláce up.
This observation is important, as it actually lends further support to Thim's suggestion that the ‘avoidance of placing particles in sentence-final position’ by seventeenth- and eighteenth-century authors may have been ‘a hypercorrect application of the normative rules against sentence-final prepositions’ (p. 235): since prepositions are typically unstressed and since sentence-final particles often are too, the two categories could easily have been targeted by the same rule.
5. Native prefixes were ousted by borrowed ones after the Norman Conquest
English verb–particle combinations are often discussed in contrast with native inseparable prefix verbs, which, as we saw above, represent an older stage of English. Only a handful of these prefixes are still used in present-day English (be-, mis-, over-, un- and under-), but in Old English there were more than a couple of dozens of them. It is well known that in the centuries following the Norman Conquest, many verbs of Romance origin were added to the Middle English lexicon. Among this influx of borrowings from French or Latin, there were numerous verbs already carrying a prefix: admit, combine, dissent, exclude, import, suppress, etc. As Thim remarks, ‘the new verbal prefixes borrowed into the language via these loanwords have traditionally been taken to seal the demise of many of the older inherited prefixes’ (p. 6).
However, Thim convincingly argues that this view is anachronistic. The majority of native Germanic prefixes had already been well on their way to disappearing since pre-Old English times. Their recession was due to such language-internal factors as progressive phonetic attrition in unaccented syllables, leading to morphological decay and ultimately loss of the affected prefix, and desemantization, whereby the prefix's originally spatial meaning gradually weakened, becoming vaguely aspectual or intensifying, functionally diffuse and in the end semantically vacuous. Thim illustrates some of this with the Germanic preverb ge-, which in Old English was a productive prefix but in Middle English underwent phonetic reduction to i- and then zero. This phonetic reduction may not necessarily have gone hand in hand with semantic bleaching, but in any case, ge- must have undergone functional loss already in Old English, unlike the retained prefix be-, which still has a marginally productive derivational use in present-day English – Thim doesn't give examples of this but one could mention bejewelled and bespectacled, as well as novel forms such as behatted, bejeaned, beketchuped, etc. Thim also points out that in present-day Dutch and German, be- still has a range of productive derivational uses, while ge- has grammaticalized as a past participle marker. He further discusses the cognate preverb ga in Gothic, a language which allows us a glimpse of an even earlier stage of prefixation in Germanic than Old English does (namely one in which preverbs could still occur independently). Already in Gothic, the meaning of ga seems to have been highly elusive. Clearly, this shows that the introduction of a new set of prefixes from Romance, though it may have overlapped in time with the decline of native prefixes, cannot have been the cause of it.
The most convincing argument against such a claim is given by Thim in a somewhat cursory remark when he deals with omission and interchangeability of bleached prefixes such as a-, ge-, be- or for-: ‘The observation that the loss or substitution of native prefixes occurs first in the Northern dialects must be seen as a strong (and probably conclusive) argument against the possibility of French influence (via the borrowing of French verbs)’ (p. 172).
6. Native prefixes were ousted by the new postverbal particles
Thim also refutes a related claim often made in the literature, namely that phrasal verbs, as a relatively new phenomenon in the English language, replaced an older pattern of verbs with Old English prefixes. This certainly cannot be correct, given that not all Old English prefixes have been lost in present-day English, as we have seen. But more importantly, Thim argues that phrasal verbs should not be seen as a (partial) replacement of a more archaic system, but as a continuation of it. Even as late as in Middle English – the period we associate with the import of Romance prefixes – new Germanic-based prefixes were still added to the English inventory, namely back-, down-, off-, in-, out- and up- (based on my comparison of Thim's table 5-2 and table 5-3). As Thim writes, ‘[t]hese younger . . . prefixes tend to be homonymous with free particles. When in the course of the Middle English period the particles stop occurring in preverbal position, their further development into bound affixes is prevented’ (p. 250, from the conclusion; see also p. 164). And he adds, ‘In all other respects, however, the particles have turned out to be fully equivalent to the prefixes, as the semantic developments of native prefixes and particles has shown’ (p. 250).
I would like to object, however, to the way this conclusion is formulated, as Thim's book does not come near to describing in any detail how prefixes and particles developed semantically. For prefixes Thim mainly refers to semantic weakening and for particles, as mentioned above, he suggests a shift from a directional to an aspectual meaning. Does Thim imply that these two processes boil down to the same thing? Any implication that aspectual particles are semantically weak, or weaker than directional particles, has to be rejected, though. While one might agree that up in, say, liven up does not have much content over and above what the simple verb already expresses by itself, this definitely does not hold for such aspectualizers as on (e.g. read on ‘continue or resume reading’), over (e.g. read it over ‘read it again’) or through (e.g. read it through ‘read it in its entirety’). Also, for the emergence of new prefixes in Middle English, it is a pity that Thim contents himself with referring to processes that receive interesting, but very schematic representations (pp. 85, 88). He never actually shows how these prefixes come into being. This book does not explain, for instance, that down (whether preverbal or postverbal) is derived from late Old English dúne or dún, weakened forms of adúne, which is itself weakened from Old English of dúne, meaning ‘off the hill or height’ (OED, s.v. down, adv.). Such information could have put some flesh on the bones of the claim that prefixes and postverbal particles develop from nouns.
This said, it is hard to find fault with Thim's point that phrasal verbs can't be seen as having supplanted an older system of prefix verbs, in view of the fact that some of the present-day particles are etymologically the same items as some of the somewhat older prefixes, the major difference between them being that they occur in different positions, as a result of the independent word-order changes mentioned before. In other words, saying that particles have replaced prefixes is a bit like saying, paradoxically, that they have replaced themselves.
7. Phrasal verbs are Germanic
It is tempting to see phrasal verbs as Germanic, especially in the light of their frequent comparison with learned Latinate verbs (give up vs abandon or surrender; put out vs extinguish, and so on). There are almost no phrasal verbs in which the verb is of Latin origin, since Latinate verbs are often prefixed (cf. supra), and the addition of a particle would therefore sound pleonastic: ?combine together, ?exclude out, ?eradicate out (p. 189).
Interestingly, though, phrasal verbs are less Germanic than one might think. Thim mentions data from one of his previous studies into the etymologies of phrasal verbs in English letters from the fifteenth and sixteenth centuries. More than a quarter of these phrasal verbs are formed with a Latin- or, more often, French-derived verb (e.g. advaunce forth, retorne back). Some of these may have gone out of use, but they show that the phrasal verb isn’t, and never was, the all-Germanic alternative to Romance words. Use up and move forward, which in the literature have been mentioned as the common counterparts of the more elevated Latinate verbs consume and advance, both contain a verb of French origin, as Thim rightly points out (p. 40). He concludes: ‘As a construction type, particle verbs turn out to be rather insensitive to the etymological origins of their components, as long as the meaning of a simple verb does not preclude its combination with spatial or aspectual particles’ (p. 195). Moreover, if it weren't for Thim's book, one would almost overlook the fact that even a couple of particles, too, have a Romance etymology. Thim mentions apart (p. 157); another French-derived particle is past.
8. Phrasal verbs used to be colloquial (but aren't anymore)
According to received opinion, phrasal verbs tend to be colloquial, even slangy or vulgar – one may think here of shut up and f— off. Thim argues that this view, still prevailing in the twentieth century, ‘has its roots in the 18th century as the indirect result of a number of metalinguistic and stylistic factors’, which he therefore describes as a ‘colloquialization conspiracy’ (p. 251, cf. pp. 233–44). Apart from the normative proscription of preposition stranding which may have led, as mentioned above, to hypercorrect action against particles, Thim also discusses style guides’ more general verdict against ending a sentence with too many monosyllables of whatever kind and their ban against ‘superfluous’ elements, as well as the phrasal verb being treated too readily as ‘idiomatic’, in the broadest possible way of being characteristically English and semantically irregular (cf. the first claim above).
Thim, drawing on earlier studies (including some of his own), argues quite extensively that there is no pre-1800 evidence whatsoever for any colloquialness of the phrasal verb. For instance, the construction wasn't more frequent in personal letters than in other genres. Also, as stated above, up to the end of the eighteenth century, there are no signs of any dislike of phrasal verbs among early grammarians and lexicographers or any avoidance of them among contemporary writers.
Thim is less clear on whether the construction is or isn't colloquial in present-day English. He writes, ‘[q]uite clearly, very many of them do belong to the more informal or ‘colloquial’ registers of the language’, but he is unwilling to draw the obvious conclusion from Biber et al.'s (Reference Biber, Johansson, Leech, Conrad and Finegan1999) findings about the occurrences of phrasal verbs across registers. He writes: ‘the overall infrequency especially of the intransitive phrasal verbs in academic prose need not be connected to their avoidance as being colloquial, but rather with the preference of formal Latinate verbs, and this would not necessarily entail the conclusion that the phrasal verbs are informal’ (p. 44). I do not think this dismissal is convincing. Surely, the choice of simple Latinate verbs over verb–particle combinations must be motivated precisely by the feeling that phrasal verbs are considered not appropriate in these registers? The observation that in the BNC (Davies Reference Davies2004–) the frequency of particles in spoken conversation is more than ten times higher than in medicine-related academic texts and still more than five times higher than in administrative writing can mean only one thing to me: the phrasal verb, as a construction, is located towards the colloquial end of the formality spectrum.
9. There is no hard evidence for an overall increase in frequency of phrasal verbs
Thim attempts to bring together quantitative data from a number of early studies to get an overview of the long-term development of phrasal verbs from the Middle Ages to the twentieth century. He does not put much faith in the outcome of his endeavours: ‘a meaningful interpretation of the figures . . . is extremely difficult’, ‘we must admit that the figures are completely inconclusive’ (p. 212) and ‘many statements made so far about the quantitative diachronic development of the phrasal verb as a construction type throughout the Modern English period are meaningless’ (p. 213).
It seems to me that Thim has wasted a lot of time and mental energy comparing these earlier studies and being frustrated with their hard-earned but divergent results. In the last couple of years, it has become possible to search a vast amount of historical language use via the Google Books project (Michel et al. Reference Michel, Shen, Aiden, Veres, Gray and Brockman2010). Figure 2 below shows the results of a search for the particles up, out, down and off, tracking their frequency in printed books since the end of the Middle English period (1500) to the year 2008, with a ten-year ‘smoothing’ of the curve. Assuming that the tagging of these items as particles is correct and consistent throughout this period (see Lin et al. (Reference Lin, Michel, Aiden, Orwant, Brockman and Petrov2012) for details on the tagger used, which achieves state-of-the-art accuracy of 97.9 per cent on the PennTreebank as a standard benchmark and 93.3 per cent accuracy on historical English from 1700 to 1914), there now appears to be little reason why Thim should write that ‘long-term changes in the frequency of occurrence of the English particle verbs are not known’ (p. 120) or why he should have hedged a statement such as ‘an overall increase between the late Middle Ages and today is quite possible’ (p. 214). Up to 1700 (the end of the Early Modern English Period), the development of the individual particles is admittedly messy and spiky (perhaps due to some scarcity in written material from that period), but the overall increase is undeniable. From then on, the evolution of these particles proceeds in sync, first for the frequently used particles up and out and from 1800 most clearly for off and down, suggesting that what is characterized by falls and rises over time here is the verb–particle construction, not individual lexical items. Note the general increase in the nineteenth century (mainly in the first half and for out especially in the second half of that century), for which even Thim finds the evidence in the literature persuasive (p. 251), and the unmistakable post-World War II decrease, which the phrasal verb has fully overcome only these last few years.
Evolution of the particles up, out, off and down in English printed books, from 1500 to 2008, via Google Books Ngram Viewer
It is a missed opportunity for a book on the history of the phrasal verb that Thim did not run such a search in Google Books Ngram Viewer or in the Corpus of Historical American English (COHA), if only briefly at the last stage of his publication project. On the other hand, Thim is right to be careful in drawing conclusions from previous quantitative studies. Likewise, the results in figure 2 are to be taken with some caution, given that automatic tagging of words as particles (as opposed to prepositions) is never fully reliable. Furthermore, data from published books may not be equally representative of the way language is used in spontaneous discourse across different periods. It is for this reason that some authors have preferred using corpora made up of private letters or diaries, which may contain more every-day language.
10. We know practically everything there is to know about phrasal verbs
Phrasal verbs are often regarded as a quintessentially English construction and English is arguably the most widely studied language there is. Therefore, if there is one single pattern which by now should be thoroughly documented and understood in virtually all its details, it has to be the English verb–particle construction. On reading Thim's book, one gets the feeling that this could not be further from the truth. Thim's critical surveying of the literature leads to an (over)abundance of statements about how little we still know about the history of phrasal verbs and related constructions. Here is a sample of such assertions:
Several details of the loss of the Old English prefixes are still in need of further exploration. (p. 164)
A more-than-cursory, in-depth study of the lexical developments of at least a representative number of particle verbs is conspicuously absent from the literature and would be highly desirable. (p. 165)
. . . the generalizations based on a critical survey of the literature which have been offered in the present section would benefit from more detailed analyses of a large number of verbs and particles. (p. 183)
. . . so far a satisfactory quantitative account of the history of the phrasal verbs remains to be written. (p. 197)
Thim is aware of the lacunae in the literature, and while one could perhaps regret that he did not actually attempt to fully address these himself, such would be more than what can be expected of a single book. At least, we now know where progress is yet to be made. The boundaries to our knowledge have meanwhile been, and will continue to be, pushed forward – see Ishizaki (Reference Ishizaki2012) and Rodríguez-Puente (Reference Rodríguez-Puente2013) for two recent diachronic studies of phrasal verbs.
By way of conclusion
Thim's book is a new and important milestone in the study of phrasal verbs, tracing its history from Proto-Indo-European and especially Germanic to present times, thus expanding a narrow focus on its wrongly assumed ‘Englishness’ to a wider cross-linguistic and diachronic recognition of its ties with other languages and with related structures, especially prefixed verbs. Sadly, though, like Spasov's (Reference Spasov1966) pioneering study critiqued by Martin (Reference Martin1990: 32), it ‘tantalizes curiosity but does not satisfy it’. Thim takes stock of the vast literature, including some studies he previously carried out himself, but does not offer a new body of empirical data that we are apparently still in need of. He rightly rebukes some of the early descriptions of the topic, because the subsequent literature too often uncritically regards these sources as reliable, ‘and in several instances the chain of quotes seems to turn into a game of Chinese whispers’ (p. 202). Occasionally, however, Thim himself misrepresents certain aspects of the present-day construction, by stating that the particle is always stressed or by treating the construction as belonging to word formation and not (also) to syntax – although the exact status of the verb–particle construction may continue to be a matter for debate.
Thim's approach is both historical and historiographical, as he not only studies the history of the construction as such but is also interested in the history of such descriptions. The book is well written, although its prose is occasionally archaic (e.g. well nigh, a goodly number) and overuses the definite article for generic plural nouns (e.g. the phrasal verbs). It is generally well organized, but Thim could have merged the section on past quantitative studies in chapter 6 with chapter 4, ‘Writing the history of the phrasal verb’; it also felt rather odd to find a broad statement such as ‘For a general account of the development of phrasal verbs and other types of complex predicates in the history of English, see Traugott (Reference Traugott, Brinton and Akimoto1999)’ as late in the book as on page 183. On the whole, the book is explicitly backward-looking, finding hiatuses in past accounts but doing little to fill them. Only in the final pages does Thim sketch new promising avenues for further research, based on his assessment of our current state of knowledge:
So far we know only little about the exact ways in which individual particles acquire their respective aspectualizing, deriviational or valency-changing properties, and even less about historical patterns of emergence of new idiomatic phrasal verbs. To me it seems that all these issues provide prime examples of constructionalization and can be discussed most fruitfully in the context of surface based exemplar models of language which do not believe any longer in the reality of boundaries between syntax, morphology and the lexicon. (pp. 252–3)
In spite of its shortcomings, Thim's book deserves praise for being a very erudite survey of previous literature on phrasal verbs and its development. The book's double thesis that phrasal verbs aren't characteristically ‘English’ and that the postverbal position of the particle is a by-product of independent changes in the language is well argued for. Also praiseworthy is Thim's attitude not to take any preconceived idea about phrasal verbs for granted, most notably received opinions about how (if indeed at all) phrasal verbs were stylistically perceived by previous generations. His close examination of early historical sources is exemplary. It is a major reference work for scholars who want to consult a critical, state-of-the-art review of past and current thinking about the verb–particle in the history of English.
