Skip to navigation – Site map

HomeIssues19PapersMotivated patterns of phrasal ver...

Papers

Motivated patterns of phrasal verbs and learner’s dictionaries

Thomai Dalpanagioti

Abstracts

Phrasal verbs have always been considered problematic for English language learners due to their polysemy and the apparently random combination of verbs and particles. Situated within the cognitive linguistic framework, the present study examines the complex interaction of lexical, semantic, syntactic and pragmatic aspects of phrasal verbs and argues for a holistic description of their phraseological and conceptual patterns, with a view to improving their representation in learner’s dictionaries. The proposed integrated approach brings together three theoretical frameworks: Corpus Pattern Analysis, Frame Semantics, and Conceptual Metaphor and Metonymy Theory. A case study of selected related phrasal verbs (walk away with, walk off with, run away with and run off with) illustrates how corpus-linguistic and cognitive-linguistic principles combine in identifying, comparing and associating phrasal verb lexical units. The resulting unified representation of the phrasal verb construction walk/run away/off with something is compared to the representation offered by monolingual learner’s dictionaries and practical suggestions for improvement are offered.

Top of page

Full text

Introduction

1This paper is motivated by the pervasive yet elusive nature of phrasal verbs, unanimously recognized as one of the most challenging aspects of the English language. Phrasal verbs (PVs) feature among the “prototypical” kinds of formulaic language (Wahl & Gries [2020: 84]) and have been defined as “multiword expression[s] consisting of a verb plus one or more particle(s)” which “may function either as an adverb (away, out) or as a preposition (with, to), or both (in, through)” (Atkins & Rundell [2008: 171]). Different approaches to PVs have examined their syntactic, semantic, cognitive or discourse features using various classification criteria such as transitivity, particle movement, degree of compositionality, conceptual motivation, corpus-based distribution (for an overview see e.g. Zhi & Juan [2015]; Rodríguez-Puente [2019]; Alangari, Jaworska & Laws [2020]).

  • 1 In this study we use the term “phrasal verb” following Thim’s [2012: 4] note that this term usually (...)

2This study addresses the issue of the representation of PVs in the context of pedagogical lexicography. PVs – also called “particle verbs” or “verb-particle constructions” –1 form a clearly identified type of phraseological units in lexicographically relevant typologies of multi-word expressions (MWEs) (see e.g. Atkins & Rundell [2008]; Bergenholtz & Gouws [2014]; Baldwin & Kim [2010]; Gantar et al. [2019]). For instance, in their overview of MWE types, Gantar et al. [2019: 147] describe PVs as “MWEs consisting of a verb plus one or more particles that show a certain degree of idiomatic meaning”. Idiomaticity may thus determine which verb-particle combinations are of lexicographic interest; yet its scalar nature can make it difficult to identify even the basic form (Gantar et al. [2019: 147-148]). The lexicographers’ task of identifying PVs in authentic data based on their semantic, syntactic and pragmatic properties and handling them systematically in dictionaries becomes even more challenging in a pedagogical context where “descriptively adequate, intuitively acceptable, and easily accessible formulations” (Taylor [1993: 220]) should be sought.

3To examine the complex interaction of lexical, semantic, syntactic and pragmatic aspects of PVs, we argue for an integrated (corpus-based and cognitively-oriented) approach which aims at a holistic description of the phraseological and conceptual patterns of PVs. Bringing together Corpus Pattern Analysis (Hanks [2013a: 404]), Frame Semantics (Fillmore [1982]) and the Conceptual Metaphor and Metonymy Theory (Lakoff & Johnson [1980]), we can reveal associations and differences among PVs. The applicability of the approach has been illustrated in the disambiguation of polysemous verbs (see e.g. Dalpanagioti [2018a, 2019]) and is explored here in relation to PVs.

4The PVs walk away with, walk off with, run away with and run off with are used as a case study. We have decided to study these PVs because they illustrate all three problem areas pointed out by Atkins & Rundell [2008: 173-175] with regard to the lexicographic treatment of PVs; namely, (1) it is not always clear if they are two- or three-part PVs; (2) as motion verb + directional particle PVs, they have a literal meaning as well as figurative meanings; and (3) they are syntactically distinct and thus physically separated in dictionaries, although they are semantically related.

5Against this background, in Section 1 we first briefly present an overview of approaches to PVs and we outline a corpus-based and cognitively informed approach to phraseology. This approach is applied in Section 2 to the analysis of the selected PVs revealing a set of parallel motivated patterns. The need for a more rational, motivated and unified lexicographic treatment becomes evident in Section 3, which reviews the corresponding entries in online versions of five English monolingual learner’s dictionaries (MLDs) and offers suggestions for improvement.

1. Background

1.1. On motion verbs and phrasal verb constructions

6Several studies have focused on motion verbs particularly in relation to the typological distinction between “manner” languages and “path” languages; English, for example, belongs to the former as it typically encodes manner of motion in the verb (e.g. walk, run, stagger), while path information appears in nonverbal elements such as particles (e.g. away, upwards, in) and prepositional phrases (e.g. across the street) (Talmy [1985, 2000]). Providing an overview of studies on human locomotion verbs, Goddard, Wierzbicka & Wong [2016: 306] observe that “research into locomotion verbs has not been greatly interested in manner-of-motion except at a macro-level” and proceed to offer a fine-grained cross-linguistic semantic/conceptual analysis of “walking” and “running”. The present study takes account of the differences in the manner of motion expressed by walk and run (mainly in foot movement pattern and speed) to analyze non-literal uses of these motion verbs in PV constructions.

7Research on English PVs has advanced from the traditional view of PVs as arbitrary verb + particle combinations to the cognitive perspective that highlights their conceptual motivation to constructionist approaches that treat them as conventional pairings of form and meaning along a continuum from lexicon to grammar (for an overview see e.g. Thim [2012]; Zhi & Juan [2015]; Jarosz [2019]). There is a major difference between the traditional non-compositional view of PVs and cognitive linguistic research that has drawn attention to the semantic relations between the component elements and their interplay within the composite structure. Summarizing the cognitive linguistic perspective on compositionality, Jarosz [2019: 31] points out that it “opposes the view that the element parts fully contribute to the meaning of the composite unit” since “there is also conceptual knowledge that is reconstructed on the basis of our background knowledge”. Various studies have analyzed the semantic contribution of the particle elements of PVs mainly by applying the image schema and conceptual metaphor theory (for an overview see Mahpeykar & Tyler [2015: 3-5]). By contrast, less attention has been paid to the contribution of the verb component; relevant studies (Mahpeykar & Tyler [2015]; Jarosz [2019]) provide further evidence for the compositional nature of PVs and account for the additional meaning that arises from the interaction of the components. Lastly, from a cognitive constructionist perspective PVs have been defined as “motivated pairings of form and meaning (constructions) embedded in semantic networks in which metaphorical meanings are motivated by more basic ones” (Torres-Martínez [2018: 279]). On this view, inheritance networks are postulated by connecting general, underspecified PV constructions to item-specific ones in a hierarchy of semantic relations.

8Situated within the cognitive linguistic framework, the present study explores the contribution of the verb and particle components of selected related PVs and provides a comparative motivated account of their meaning and use. PVs are viewed as constructions, but they are not analyzed in terms of Construction Grammar; instead, we apply its “sister theory”, Frame Semantics, which “takes an alternative view of constructions by first discussing the meaning side of constructions” as opposed to most research in Construction Grammar which “typically emphasizes form over meaning” (Boas [2021: 44]). In addition to the Goldberg-type abstract meaningful constructions, Boas [2013: 191] has proposed more concrete constructions: “individual verb senses should be regarded as mini-constructions with their own frame semantic, pragmatic, and syntactic specifications whenever abstract meaningful constructions overgenerate”. In line with this lexical-constructional view, we aim to identify these types of information following the methodology outlined in the next section.

1.2. A corpus-based and cognitively informed approach to phraseology

9This section briefly reviews the three theoretical frameworks that inform the present study, namely Corpus Pattern Analysis, Frame Semantics and Conceptual Metaphor and Metonymy Theory. These approaches have been combined in establishing, describing and distinguishing lexical units (both single-word and multi-word ones) for a pre-lexicographic database for manner-of-motion verbs in English and Modern Greek (Dalpanagioti [2013, 2018a]) and in reconstructing entries for MLDs (Dalpanagioti [2018b, 2019]). Lexical Units (LUs), i.e. words in one of their senses, are constructions too; as Boas [2020: 289] explains, LUs are “(lexical) constructions whose form pole is one or more word-forms, and whose meaning pole is usually represented as a specific semantic frame”. This study provides an illustrative example of how corpus-linguistic and cognitive-linguistic principles combine in identifying and comparing phrasal verb lexical units (PV LUs) and discusses the lexicographic implications of the analysis.

  • 2 CPA analysis of PVs like blow up and wipe out is further illustrated in Hanks [2013b: 729; 2020: 30 (...)

10The first step to this end is retrieving frequent patterns of co-occurrence from corpus data. The power of the corpus to foreground the syntagmatic aspect of lexis has led to a contextual theory of meaning expressed in statements such as “every distinct sense of a word is associated with a distinction in form” (Sinclair [1987: 89]), “context disambiguates” (Moon [1987: 87]), and “context determines meaning” (Hanks [2020: 297]). Implementing this idea, Hanks has proposed a corpus-driven technique for associating word meaning with word use, i.e. Corpus Pattern Analysis (CPA), which is theoretically supported by the Theory of Norms and Exploitations (Hanks [2004, 2013a]) and applied in the compilation of the Pattern Dictionary of English Verbs (PDEV). In CPA meanings are not associated with words in isolation but rather with lexico-grammatical patterns typically observed in concordance lines (Hanks [2020: 298]). In PDEV each clause role in a pattern is populated by a lexical set, i.e. a set of collocates which share a semantic meaning. To illustrate how CPA patterns can help us distinguish senses, Figure 1 shows three patterns recorded in PDEV for the PV blow up.2 Lexical sets are usually assigned a semantic type label, enclosed in double square brackets, while curly brackets represent specific lexical sets which cannot be grouped into a semantic type. As the last two patterns in Figure 1 indicate, “very often different meanings of a verb are distinguished only by the semantic types of its arguments” (Hanks [2013b: 729]).

Figure 1. CPA patterns taken from the PDEV entry for blow

Figure 1. CPA patterns taken from the PDEV entry for blow
  • 3 Core FEs are the conceptually necessary components of a frame; indications of time, place and manne (...)
  • 4 FrameNet uses font and colour differences for the names of frames and FEs. The present work uses sm (...)

11The present study combines CPA and Frame Semantics (FS), which are considered complementary approaches, as “each CPA pattern can in principle be plugged into a FrameNet semantic frame” [Hanks 2021]. Both PDEV (a CPA product) and FrameNet (an application of FS) seek to “identify stereotypical contexts”; the former focuses on the “phraseological context”, while the latter concentrates on “the context of situation in which words are used” [Hanks 2013b: 729]. The main assumption of FS is that words must be grouped and explained in relation to a “(semantic) frame”, i.e. a structured background of experience which constitutes a kind of prerequisite for understanding the meaning of a word [Fillmore 1985: 224]. Every semantic frame consists of specific “frame elements” (FEs), i.e. the “various participants, props, and other conceptual roles” involved in the schematic representation of a situation [Fillmore & Petruck 2003: 359]. Applying FS to lexicography, the Berkeley FrameNet project develops frame descriptions, establishes LUs as annotation targets, extracts sentences from a corpus (the BNC), and annotates them with frame-relevant labels, phrase types, and grammatical functions (Fillmore & Baker [2010: 320-333]; Ruppenhofer et al. [2016: 7-8]; Boas [2017: 553-564]). Each frame report defines a frame with its core and non-core FEs,3 provides sentences exemplifying prototypical instances of the FEs, points out frame-to-frame relations, and lists frame-evoking LUs. By way of illustration, Figure  2 presents part of the report for the [Destroying] frame, which is evoked by single-word LUs (e.g. destroy, devastate, ruin) and multi-word ones (e.g. blow up, take out).4 In fact, the [Destroying] LU of blow up could be linked to the first CPA pattern shown in Figure 1.

  • 5 For information on the other types of null instantiation that FrameNet recognizes but do not appear (...)

12FrameNet specifies valence (i.e. the constituents with which words combine in grammatical sentences) in both semantic and syntactic terms by linking FEs to their syntactic realizations (grammatical functions and phrase types). Since in our case study we use FrameNet’s annotation labels, these are briefly presented here and illustrated through the (partial) valence table for the [Destroying] LU of blow up in Figure 3. Phrase type tags include NP (Noun Phrase), PP (Prepositional Phrase), VP (Verb Phrase), AVP (Adverb Phrase), etc., while Ext (External Argument), Obj (Object) and Dep (Dependent) are grammatical function tags. The labels Ext and Obj are assigned to sentence constituents that fill core syntactic slots, i.e. function as subject and object respectively; “all other constituents accompanying a syntactic head are considered dependents given that their presence in a construction centred on the head is licensed by the head” (Ruppenhofer et al. [2016: 69]). Lastly, DNI (Definite Null Instantiation) marks a core FE which “is not explicitly expressed, but specific information about its interpretation is available from context” (Atkins, Fillmore & Johnson [2003: 269]); in other words, the omitted element is anaphorically understood (Ruppenhofer et al. [2016: 28]).5

Figure 2. Part of FrameNet’s [Destroying] frame report

Figure 2. Part of FrameNet’s [Destroying] frame report

Figure 3. Part of FrameNet’s valence table for blow up as a [Destroying] frame-evoking LU

Number Annotated

Patterns

8 TOTAL

Destroyer

Patient

(6)

NP
Ext

NP
Obj

(2)

PP[by]
Dep

NP
Ext

2 TOTAL

Destroyer

Patient

Place

(1)

DNI
--

NP
Obj

PP[on]
Dep

(1)

NP
Ext

NP
Obj

PP[in]
Dep

13If we consider the frame-evoking LUs listed under the [Destroying] frame we will notice that some LUs evoke the frame literally (e.g. destroy) while others non-literally (e.g. blow up). However, this information is not recorded in FrameNet, since, as Ruppenhofer et al. [2016: 100] note, “only a few metaphor relations have been added to the database”; FrameNet has not “annotated both the source and target domains on the same sentence, since such work is worthy of an entire research project in itself” (Ruppenhofer et al. [2016: 101]). Since our case study is interested in the motivation of PVs, we additionally take account of another frame-semantic project which addresses the special issue of metaphor representation. Within the framework of the SALSA (SAarbrücken Lexical Semantics Annotation and Analysis) project, which is developing frame-semantic resources for German, a double annotation scheme has been proposed with “a source frame representing the literal meaning, and a target frame representing the figurative meaning” (Burchardt et al. [2009: 216]). Double frame-semantic annotation is used in the present work to capture both the overall meaning (target frame) and the internal structure (source frame) of the PVs under study. Relevant in this respect are the conceptual mechanisms of metaphor and metonymy which can account for the link between the frames involved and reveal the motivation behind PV constructions.

  • 6 For example, Ruiz de Mendoza & Galera [2011: 20] analyze the PV burst into (as in “He burst into te (...)

14Within cognitive linguistics metaphor and metonymy are viewed as phenomena fundamental to the structure of the conceptual system rather than mere stylistic features of language. State-of-the-art overviews of Conceptual Metaphor Theory (Kövecses [2021]) and Conceptual Metonymy Theory (Ruiz de Mendoza [2021]) summarize the stages in their development from Lakoff & Johnson’s [1980] initial proposal to more fine-grained typologies and extended frameworks in the light of contextual factors. The focus of discussion seems to shift from traditional distinctions – in terms of mapping (across two separate domains in metaphor vs. within a single domain in metonymy), function of the conceptual relationship (imagistic reasoning in metaphor vs. shift of reference in metonymy) and nature of the conceptual relationship (similarity in metaphor vs. contiguity in metonymy) – to investigation of their interaction patterns. For example, Ruiz de Mendoza & Galera [2011] discuss different types of metaphor-metonymy interaction in relation to PVs and they conclude that it is not enough to combine verbal meaning with the image schematic meaning of particles, rather we may need to combine “two metaphors (which in turn may include cases of metonymic activation) either in the form of amalgams or chains” (Ruiz de Mendoza & Galera [2011: 23-24]).6 Similarly, we use metaphor and metonymy as tools for shedding light on the interpretation of the PVs under study.

  • 7 This semantic order approach, which argues that the core meaning should precede extended uses, stan (...)

15The implications of Conceptual Metaphor and Metonymy Theory (CMMT) for pedagogical lexicography are mostly discussed in relation to ordering and defining senses of polysemous headwords. For instance, Van der Meer [1999] argues that making learners aware of the extensions of words, by ordering senses in the dictionary from literal to figurative, can facilitate vocabulary learning.7 Similarly, it is important to show the relation between senses in the wording of definitions; as Lew [2013: 299] explains, “foregrounding the links between different shades of meaning may help repair some of the damage done by artificially chopping semantic space into separate dictionary senses”. Lexicographic applications of the CMMT to the treatment of MWEs can be traced in specialized dictionaries for phrasal verbs or idioms, which seek to express the underlying conceptual motivation (for an overview see Kövecses & Csábi, [2014: 129-130]), and in the “metaphor boxes” of the MEDAL (print and electronic) dictionaries (for an overview see [Moon 2004]). Complementing the linear organization of PV entries with semantic networks is another suggestion in the direction of making entries cognitively informed and user-friendly (Perdek [2010: 1394-1396]).

16Having explained how the three theoretical frameworks, i.e. CPA, FS and CMMT, are linked together, we proceed to apply the corpus-based and cognitively informed approach to the analysis of the PV construction walk/run away/off with something (Section 2) with a view to comparing the resulting unified representation to the representation offered by MLDs (Section 3). At this point we should also note that currently PDEV does not include an entry for walk or run, while FrameNet includes one walk-phrasal verb LU and four run-phrasal verb LUs but not the ones we study here.

2. Identifying motivated patterns of phrasal verbs: the case of walk/run away/off with something

2.1. Frame semantic-syntactic analysis of corpus data

17The data for the study is drawn from two corpora, i.e. the BNC and the ukWaC, accessed through the Sketch Engine interface. The combined use of these corpora provides a more representative basis for the analysis, as they differ in size (21,149 vs. 203,991 walk occurrences and 40,876 vs. 608,029 run occurrences, respectively) and coverage of text types. As Ferraresi et al. [2008: 7] point out, the BNC has a higher proportion of narrative fiction texts and spoken texts, and is characterized by a stronger historical perspective, whereas the ukWaC contains comparatively more texts dealing with the Web, education and public sphere issues, and is characterized by a stronger concern with the present time.

  • 8 For a similar discussion of the “verb + particle” combination as a compositional construction and a (...)

18The Corpus Query Language function of Sketch Engine enables us to make targeted searches for concordances with the MWEs under study, i.e. the three-part inseparable PVs walk away with, walk off with, run away with and run off with. Table 1 thus shows the number of occurrences of the target combinations extracted from the two corpora. However, not all instances of the “verb + adverbial particle + prepositional particle” combination should be regarded as PV; if, for example, we compare “Robin then started walking away with the small dog in his arms.” with “These companies will be lucky to walk away with a few billion.” we realize that the combination in the first example is compositional (i.e. not a phrasal verb).8 We therefore need to focus on the non-compositional uses of the target combinations. The Word Sketch (Multiword Sketch in particular) function of Sketch Engine helps us make some preliminary observations about the typical contexts in which the target combinations appear; for example, we notice that nouns related to championship and money appear in the with- prepositional phrase of all four target combinations, whereas imagination seems to co-occur only with run away with. Word Sketches are hence used as a springboard for a thorough analysis of concordances; they make it easier to identify separate LUs when scanning the corpus examples.

Table 1. Corpus data under study

BNC

ukWaC

walk away with

43

1,031

walk off with

39

320

run away with

147

862

run off with

123

576

TOTAL

352

2,789

19Separate LUs generally correspond to different frames and assign different FEs (Atkins [2008: 256-257]; Atkins, Rundell & Sato [2003: 335-337]). Corpus examples are thus clustered together based on the semantic frame evoked by the PVs under study; Table 2 presents sample corpus sentences annotated with core FE labels. What is striking is that all four PVs can evoke the [Win_prize] frame and the [Theft] frame in appropriate contexts, while run away with can additionally evoke the [Control] frame. These frames are not associated with the [Self_motion] frame prototypically evoked by the core verbs walk and run through the frame-to-frame relations recorded in FrameNet (i.e. Inheritance, Subframe, Causative of, Inchoative of, Perspective on, and Using, see Ruppenhofer et al. [2016: 79]). To demonstrate how the LUs are linked in a polysemous construction, we first examine the syntactic realization of the FEs and then their lexical realization.

Table 2. Assigning semantic frames to corpus examples of walk/run away/off with something

  • 9 Frame definition: “A competitor claims a prize as a result of the outcome of their participation in (...)
  • 10 Frame definition: “These are words describing situations in which a perpetrator takes goods from a (...)
  • 11 Frame definition: “A controlling_entity, controlling_situation, or controlling_variable control a d (...)

1. Frame: [Win_prize]9

a

(1) Manager Lennie Lawrence said: A few weeks ago I forecast Blackburn Roverscompetitor would walk away with the championshipprize.

(2) If there was ever a real fight between her and Damian Flint, he would take supremacy, she would be overpowered by the fierce excitement burning in her now, and hecompetitor would walk away with victoryprize on a scale she did not dare imagine.

b

(3) At this year’s Chelsea Flower show, a group of prisonerscompetitor walked off with a silver medalprize for a garden display.

(4) They were so enthusiastic for these that its pupilscompetitor regularly walked off with all the trophiesprize on sports days.

(5) It seems as if the top 10 firmscompetitor are walking off with all the prizesprize these days – literally.

c

(6) Ipswich Towncompetitor could run away with the titleprize but there is strong competition for the second spot.

(7) Lawrence said: ‘I said ten days ago Blackburncompetitor would run away with the championshipprize’.

(8) But Manorcompetitor ran away with the matchprize in the second half, scoring seven more goals.

d

(9) It was hardly likely to be repeated in 1977, when Nikicompetitor returned to form with a vengeance and ran off with the championshipprize by a huge margin, leaving Master James stranded in fifth place.

(10) Quite the reverse for Valenciacompetitor, who completed their best season ever, running off with the league titleprize and the Uefa Cup, with Vicente superb on the left wing.

2. Frame: [Theft]10

a

(11) A thiefperpetrator walked away with a haulgoods worth hundreds of pounds when he staged a ram raid on a shop with a wheelbarrow.

(12) But Keyser Ullmann had to be rescued following the failure of a £17 million corporate loan to someoneperpetrator with no references who walked away with the moneygoods.

b

(13) Then there’s the lending it out problem, where presumably people say ‘can I borrow this stuff’ and theyperpetrator just sort of walk off with itgoods and you don’t know whether you’ve got it all back, and you find a few weeks later there’s a lead missing.

(14) ‘It’s mine and we just saw the dogperpetrator walking off with itgoods’ the golfer replied.

(15) One can only guess at how Howard and Redwood must feel about taking over a department, only to find that one of their political opponentsperpetrator has walked off with the moneygoods.

c

(16) She commented afterwards that she had made no money on the venture: ‘I lost nothing, neither did I gain much, othersperpetrator run away with the profitgoods’.

(17) We believe that Council Tax should be fair and are determined to prevent local authoritiesperpetrator from running away with people’svictim moneygoods.

d

(18) Krystleperpetrator in turn cons Frank, running off with hisvictim moneygoods.

(19) A pickpocketperpetrator running off with the victim’svictim wallet and moneygoods.

(20) The menperpetrator then ran off with £513goods and escaped in a waiting car, an Austin Allegro.

3. Frame: [Control]11

(21) She stopped, suddenly realising that she was letting her ideascontrolling_entity run away with herdependent_entity.

(22) I’m afraid your rather fervid imaginationcontrolling_entity is running away with youdependent_entity.

(23) Robbie’s voice faltered into silence, and she held her breath, realising that her wretched curiositycontrolling_entity had run away with herdependent_entity again.

(24) She stopped, aware that her tonguecontrolling_entity was running away with herdependent_entity in the heat of the moment.

(25) ‘It’s a dangerous thing to let your tonguecontrolling_entity run away with youdependent_entity,’ she told him.

(26) Don’t run away with the ideacontrolling_entity that anybody’s ever helped us.

(27) Although a very great proportion of the revival in kite flying is attributed to the growth in popularity of the ‘Stunt’ kite, onedependent_entity should not run away with the ideacontrolling_entity that controlled flight is a new discovery.

(28) A lot of peopledependent_entity ran away with the ideacontrolling_entity that they were Pacifists, but so far as he was concerned that was not true.

(29) But I mustn’t let youdependent_entity run away with the impressioncontrolling_entity that I spent all my time lounging around the pool.

(30) ‘Oh, don’t run away with the impressioncontrolling_entity that we’re not concerned,’ says the official defensively.

20Applying the double annotation scheme outlined in Section 1.2., in Table 3 we provide the semantic-syntactic valence patterns of the LUs in terms of source and target frames. The target frames (namely, [Win_prize], [Theft], [Control]) describe the meaning actually conveyed by the phrasal LUs, while the source frame ([Self_motion] in all cases) captures their internal structure. Evidently, there is no direct correspondence between source/target FEs and arguments, but rather in all cases two source FEs, i.e. source (realized by the adverbial particles away or off) and cotheme (realized by the prepositional particle with), are incorporated into the frame-evoking element of the target frame. In this way, the source structure “walk/run + Adv- away/off + PP- with” is mapped onto a target transitive structure with specific contextual properties (see Section 2.2.) and conceptual motivation (see Section 2.3.).

Table 3. Double annotation of PV LUs

PVs

Frames

Valence patterns

walk away with sth

walk off with sth

run away with sth

run off with sth

Target: [Win_prize]

[competitor/ Ext/ NP]

[prize/ Obj/ NP]

Source: [Self_motion]

[self_mover/ Ext/ NP]

[source/ Dep/ Adv- away/ off]

[cotheme/ Dep/ PP- with]

walk away with sth

walk off with sth

run away with sth

run off with sth

Target: [Theft]

[perpetrator/ Ext/ NP]

[goods/ Obj/ NP]

Source: [Self_motion]

[self_mover/ Ext/ NP]

[source/ Dep/ Adv- away/ off]

[cotheme/ Dep/ PP- with]

run away with sth

Target: [Control]

a (see examples 21-25)

[controlling_entity/ Ext/ NP] [dependent_entity/ Obj/ NP]

b (see examples 26-30)

[dependent_entity/ Ext/ NP or CNI]

[controlling_entity/ Obj/ NP]

Source: [Self_motion]

a

[self_mover/ Ext/ NP]

[source/ Dep/ AVP- away]

[cotheme/ Dep/ PP- with]

b

[cotheme/ Ext/ NP or CNI]

[source/ Dep/ AVP- away] [self_mover/ Dep/ PP- with]

2.2. Phraseological pattern analysis

21In this section we combine CPA with FrameNet features to describe the phraseological patterns of the PV LUs. Following CPA, we specify the semantic type of the PVs’ arguments and using FrameNet we associate the semantic types of argument fillers with FEs. Table 4 shows each PV’s semantic preference, which determines the range of nouns normally found in the clause roles, semantic prosody, which expresses the associated evaluative/attitudinal meaning, and formality level. Both similarities and differences between the phraseological patterns of the PV LUs can be seen in Table 4.

22There is a striking similarity between the patterns of the [Win_prize] and [Theft] frame-evoking LUs, which are differentiated mainly by the lexical items that appear to be the normal collocates in the object slot. Although in both cases the PVs have the same implication (i.e. take something easily), sound informal and collocate with nouns which denote something valuable, the most typical collocate of the [Win_prize] frame-evoking LUs is championship while the [Theft] LUs collocate with amounts of money and related lexical items.

  • 12 Only one relevant instance has been found in the BNC with off instead of away: “It was heartfelt an (...)

23On the contrary, the phraseology of the [Control] frame-evoking LU is quite distinct. First of all, only one combination (i.e. run + away + with) evokes this frame.12 In addition, the FEs of the two relevant valence patterns (see Table 3) are realized by a smaller set of items and the resulting phraseological patterns take two typical forms (see Table 4). In the first pattern the controlling_entity FE appears in the subject slot and is realized by lexical items that denote mental entities such as thoughts and feelings (semantic preference), while the dependent_entity FE appears in the object slot and is realized by items from the grammatical category of personal pronouns (colligation). The second pattern is more restricted and seems to be an MWE characterized by negative semantic prosody (not) and a very limited choice from the semantic class of mental entities (idea/impression) in the object slot where the controlling_entity FE appears; the dependent_entity FE is often the personal pronoun you, which can be constructionally null instantiated because of the imperative structure in the context of a spoken conversation.

24The combination of conceptual (FS) and contextual (CPA) terms to represent the patterns of the PVs under study may have already provided clues to the motivation of these patterns but this issue is clearly addressed in the next section, which explains why the PVs under study should not be treated separately but together as a polysemous construction.

Table 4. Contextual specification of PV LUs

PVs

Frames

Phraseological patterns

walk away with sth

walk off with sth

run away with sth

run off with sth

[Win_prize]

competitor collocate type: human

prize collocate type: something valuable (e.g. championship, victory, trophy, medal, title)

semantic prosody: it implies that you win in a relaxed, confident way almost making fun of your competitor

register: informal

walk away with sth

walk off with sth

run away with sth

run off with sth

[Theft]

perpetrator collocate type: human

goods collocate type: something valuable (e.g. money, profit, bag, wallet)

semantic prosody: it implies that you steal in a relaxed, confident way almost making fun of the victim

register: informal

run away with sth

[Control]

controlling_entity collocate type: thoughts/feelings (e.g. imagination, idea, curiosity, desire, impression, tongue)

dependent_entity collocate type: human

typical patterns:

[ideas/feelings] run away with [pronoun]

not run away with the idea/impression that

semantic prosody: it implies disapproval due to lack of mental control

2.3. A unified representation of walk/run away/off with something

25Having presented the similarities and differences between the PV LUs under study in terms of semantic-syntactic and phraseological patterns, we argue in this section that we are dealing with a single polysemous and partially filled construction rather than with distinct PVs. We thus consider how the PV LUs are motivated by the cognitive mechanisms of metonymy and metaphor. In line with Mahpeykar & Tyler [2015] and Jarosz [2019], we point out the contribution of each component to the overall meaning as well as the additional meaning contributed by the interaction of metonymy and metaphor.

  • 13 Frame definition: “The self_mover, a living being, moves under its own direction along a path. Alte (...)
  • 14 Frame definition: “An object (the theme) moves away from a source. The source may be expressed or i (...)

26Table 5 shows how meaning is built cumulatively by gradually adding the components at the top of the Table, while at the bottom of the Table we see the kind of contribution that each component seems to make. The manner-of-motion verbs walk and run prototypically evoke the [Self_motion] frame,13 but when they combine with the particle away or off a sub-frame of the motion scenario is evoked, i.e. [Departing],14 which profiling the source FE expresses some change of location (physical separation). When the prepositional particle with is added to the previous combination, the cotheme FE is profiled, which denotes a second moving object that moves away along with the self_mover. So far the combination is compositional and is not considered a PV (as also explained in Section 2.1.). It is at this point that conceptual mechanisms come into play triggering a set of figurative PV LUs.

  • 15 Frame definition: “A recipient starts off without the theme in their possession, and then comes to (...)

27The link between the source [Self_motion] frame and the target frames presented above in Tables 2, 3 and 4 is the metonymic activation of the [Getting] frame.15 More precisely, when the source cotheme is inanimate (or construed as inanimate), the PV construction walk/run away/off with something acquires the general meaning “get something” via the metonymy action (i.e. moving away with something) for result (i.e. obtaining it). The contextual specification of the theme of possession (in the with- slot) as prize or goods gives rise to the more specific frames [Win_prize] or [Theft] respectively; the source from which the theme came is not profiled; it is underspecified in the PV construction as away or off, but is understood from the context as a competition or an owner respectively. The implication of the [Win_prize] and [Theft] frame-evoking LUs, i.e. obtaining the prize or the goods easily, can be accounted for based on the manner expressed by the motion verbs walk (i.e. effortlessly) and run (i.e. quickly).

  • 16 For an analysis of the self metaphors, see Kövecses [2005: 55-56].
  • 17 It is thus clear why we treat the pattern [ideas/ feelings] run away with [personal pronoun] under (...)

28When the theme of the underspecified [Getting] frame is realized by a non-literal object of possession, the [Control] frame is evoked through a metaphorical procedure that imposes a number of selectional restrictions on the general PV construction. The metaphor underlying the relevant phraseological patterns is (lack of) self-control is (not) being in one’s normal location;16 what controls the self (forcing it to change its location) is a mental entity (see the controlling_entity collocate type in Table 4). Therefore, the metaphor ideas/ feelings are moving objects motivates both the MWE (not) run away with the idea/impression that and the pattern [ideas/feelings] run away with [personal pronoun], while the latter is additionally motivated by the metaphor human is object.17 The implication of the [Control] LU is negative, i.e. ideas/feelings control one’s self making them say/do something stupid or making them believe that something is true when it is not. This LU thus takes only run in the verb slot because the fast manner-of-motion it expresses is compatible with the forceful control action. As regards the restriction on the particle slot, it seems to be a matter of frequency rather than conceptual motivation; away is conventionally used, and, although off is not conceptually incompatible either, it appears rather infrequently in this context.

29On the whole, a unified treatment of the walk/run away/off with something construction is possible if we consider the conceptual mechanisms that link the PV LUs together and motivate their phraseological patterns. For instance, an interesting case of metaphor metonymy interaction is found in the example “It’s a dangerous thing to let your tongue run away with you.”. The controlling_entity slot is not filled directly with a lexical item from the semantic class of mental entities (see Table 4); instead, tongue (speech organ) metonymically stands for the words spoken, which in turn stand for the ideas expressed, which are metaphorically construed as the moving entities controlling an objectified, mentally weak human being.

Table 5. Accounting for the meaning and use of walk/run away/off with something

Table 5. Accounting for the meaning and use of walk/run away/off with something

3. Representation of phrasal verbs in online MLDs

3.1. Dictionary review

30In this section we examine how the walk/run away/off with something construction is treated in the online editions of CALD, COBUILD, LDOCE, MEDAL and OALD. The first observation we need to make is that all dictionary entries provide four separate entries (with the headwords walk away with something, walk off with something, run away with something, and run off with something) with no link between them. Therefore, users are not provided with a unified picture of the PV LUs and phraseological patterns analyzed above as a single construction. Table 6 presents an overview of the LUs recorded in the entries of the five online MLDs.

31As regards the coverage of the LUs, the [Win_prize] sense is recorded in all dictionaries under three of the four PV entries; it does not appear under the run off with something entry although our corpus study provides evidence for it (for sample examples see Table 2, 1d). Similarly, the [Theft] sense is not recorded under the walk away with something entry or the run away with something entry, with the notable exception of MEDAL. The [Control] sense can be found in all dictionaries, but there is considerable variation in the presentation of the relevant phraseological patterns; they might appear as sub-entries highlighting specific patterns (see e.g. LDOCE 2, 3 and MEDAL 4), within definitions (see e.g. COBUILD 1, 3), or as separate entries (see e.g. CALD and OALD).

32The implications and stylistic markedness of the PV LUs are captured in most entries through the definitions and parenthetical notes respectively. However, the entries show no particular regard for a rational organization of the sense divisions. If, for instance, we examine the run away with something entries, we see that phraseological patterns of the same LU are distanced from each other (see e.g. COBUILD 1, 3 and MEDAL 1, 4), separate LUs do not appear in a rational order (e.g. from literal to metaphorical possession as shown in Table 5) (see e.g. LDOCE and MEDAL), and definitions do not seem to include clues to conceptual links either.

Table 6. PV LU coverage in the “Big Five” MLDs

PV

MLD

walk away with something

walk off with something

run away with something

run off with something

CALD

to win a prize or competition very easily

1. to win something easily

2. to steal something or take something without asking permission

to win a competition or prize very easily

Separate entry: run away with sb: If a feeling or idea runs away with you, you cannot control it and it makes you behave stupidly.

to borrow, steal, or take something that does not belong to you

COBUILD

If you walk away with something such as a prize, you win it or get it very easily.

(journalism)

1. If someone walks off with something that does not belong to them, they take it without permission.

(informal)

2. If you walk off with something such as a prize, you win it or get it very easily.

(journalism)

1. If you let your imagination or your emotions run away with you, you fail to control them and cannot think sensibly.

2. If someone runs away with a competition, race, or prize, they win it easily.

3. If you run away with a particular idea, you accept it without thinking about it carefully, even though it is wrong.

to steal

LDOCE

to win something easily

(informal)

1. to win something easily

(informal)

2. to steal something or take something that does not belong to you

(informal)

1. run away with you: if your feelings, ideas etc run away with you, they start to control how you behave

2. your tongue runs away with you: if your tongue runs away with you, you say something that you did not intend to say

3. run away with the idea/impression (that) to think that something is true when it is not (spoken)

4. to win a competition or sports game very easily

(informal)

to steal something and go away

(informal)

MEDAL

1. to feel a particular emotion when you leave a situation

2. to win something easily

3. to steal something

1. to steal something

2. to win something easily

1. if feelings, ideas etc run away with someone, they make someone say or do something stupid

2. to steal something, or to borrow something without asking

(informal)

3. to win a competition, game, or prize very easily

4. run away with the idea/impression that: to believe that something is true when it is not

to steal something or to take it without permission

(informal)

OALD

to win or obtain something easily

(informal)

1. to win something easily

2. to take something that is not yours; to steal something

(informal)

1. to win something clearly or easily

2. to believe something that is not true

Separate entry: run away with you: if a feeling runs away with you, it gets out of your control

to steal something and take it away

3.2. Implications of the corpus-based and cognitively informed analysis

33A number of suggestions can be put forward if we consider the dictionary entries just examined in the light of the unified representation of the walk/run away/off with something construction presented in Section 2. On the one hand, our targeted corpus investigation has revealed that the [Win_prize] and [Theft] senses should be consistently associated with all four PV forms (walk away with, walk off with, run away with and run off with). On the other hand, our cognitively informed analysis has made it clear that there is a non-arbitrary relationship between the PV LUs and that their phraseological patterns are motivated. However, the underlying conceptual motivation is not reflected in current dictionary practices, which create distance between semantically related phraseological patterns and do not help users (learners) see the relation among senses and uses. We thus propose cognitively oriented dictionary features to improve the macro- and micro-structure of entries.

34Firstly, taking advantage of the flexibility of the electronic medium, we can make a single entry with the headword walk/run away/off with something accessible through all relevant PV forms. In a reorganized entry in Figure 4, we use frame-based signposts as guidewords in combination with a tiered structure and we clearly show the phraseological patterns related to each sense. Parallel wording in guidewords or definitions can also make connections more transparent (e.g. “take something easily” and “take control”). As there are no space constraints in electronic dictionaries, awareness-raising notes about the underlying motivation of phraseological patterns can be provided as hyperlinks, extending thus MEDAL’s practice of metaphor boxes to cover more uses.

35Further investigation into related combinations (e.g. walk / run / other motion verb? + away/off + somebody) can reveal a broader network of related PVs and potentially a more general motivated construction.

Figure 4. A reconstructed entry for the PV construction

Figure 4. A reconstructed entry for the PV construction

Conclusion

36This paper has outlined and demonstrated a corpus-based and cognitively informed approach to the representation of PVs with implications for the compilation of learners’ dictionaries. More precisely, we have explored the compositional nature of selected related PVs and provided a comparative motivated account of their meaning and use by demonstrating all the steps involved in the process (i.e. retrieving corpus data, assigning semantic frames to corpus examples, annotating valence patterns in frame semantic-syntactic terms, identifying distinct phraseological patterns for each PV LU, tracing the underlying conceptual motivation), thus eventually understanding what the components of the PVs contribute to the overall meaning and how the additional meaning emerges. The detailed case study of the walk/run away/off with something construction has shown that by foregrounding conceptual and phraseological similarities and differences among PVs, we can enrich existing online dictionary entries for PVs, which are treated in a disconnected manner (separately from the main verb entry and from other related PVs, usually with no links between related senses). Although one could say that to apply such a procedure for all PVs listed in MLDs would be so time- and labour-consuming as to make it hardly feasible, the systematic analysis and representation of the motivation underlying a set of the most frequent PVs would be a realistic and promising project in EFL lexicography. Besides, the proposed integrated framework is in line with the interdisciplinary, team-based nature of lexicographic work.

37Considering Lakoff’s [1987: 346] claim that “it is easier to learn something that is motivated than something that is arbitrary”, the proposed approach to the description and disambiguation of PV LUs can be useful not only in the context of pedagogical lexicography but of EFL teaching as well (e.g. in designing Data-Driven Learning tasks for raising awareness of motivated patterns in PVs). Future studies should investigate the effectiveness of pedagogical materials developed along these lines and address the question whether and to what extent dictionary users would appreciate or profit from such representation of PVs.

Top of page

Bibliography

Alangari Manal, Jaworska Sylvia & Laws Jacqueline, 2020, “Who’s afraid of phrasal verbs? The use of phrasal verbs in expert academic writing in the discipline of linguistics”, Journal of English for Academic Purposes 43, 1475-1585.

Atkins Sue, 2008, “Then and now: Competence and performance in 35 years of lexicography”, in Fontenelle Thierry (Ed.), Practical lexicography: A reader, Oxford: Oxford University Press, 247-272.

Atkins Sue, Fillmore Charles & Johnson Christopher, 2003, “Lexicographic relevance: Selecting information from corpus evidence”, International Journal of Lexicography 16(3), 251-280.

Atkins Sue & Rundell Michael, 2008, The Oxford Guide to Practical Lexicography, Oxford: Oxford University Press.

Atkins Sue, Rundell Michael & Sato Hiroaki, 2003, “The contribution of FrameNet to practical lexicography”, International Journal of Lexicography 16(3), 333-357.

Baldwin Timothy & Kim Su Nam, 2010, “Multiword expressions”, in Indurkhya Nitin & Damerau Fred (Eds.), Handbook of Natural Language Processing, 2nd Edition, Boca Raton, FL: CRC Press, 267-292.

Bergenholtz, Henning & Gouws Rufus, 2014, “A Lexicographical Perspective on the Classification of Multiword Combinations”, International Journal of Lexicography 27(1), 1-24.

Boas Hans, 2013, “Cognitive construction grammar”, in Hoffmann Thomas & Trousdale Graeme (Eds.), The Oxford Handbook of Construction Grammar, Oxford: Oxford University Press, 233-254.

Boas Hans, 2017, “Computational resources: FrameNet and Constructicon”, in Dancygier Barbara (Ed.), The Cambridge Handbook of Cognitive Linguistics, Cambridge: Cambridge University Press, 549-573.

Boas Hans, 2020, “Constructions in English grammar”, in Aarts Bas, McMahon April & Hinrichs Lars (Eds.), The Handbook of English Linguistics, Oxford: Wiley, 277-297.

Boas Hans, 2021, “Construction grammar and frame semantics”, in Wen Xu & Taylor John (Eds.), The Routledge Handbook of Cognitive Linguistics, New York/London: Routledge, 43-77.

Burchardt Aljoscha, Erk Katrin, Frank Anette, Kowalski Andrea, Padó Sebastian & Pinkal Manfred, 2009, “FrameNet for the semantic analysis of German: Annotation, representation and automation”, in Boas Hans (Ed.), Multilingual FrameNets in Computational Lexicography: Methods and Applications, Berlin: Mouton de Gruyter, 209-244.

Dalpanagioti Thomai, 2013, “Frame-semantic issues in building a bilingual lexicographic resource: A case study of Greek and English motion verbs”, Constructions and Frames 5(1), 5-38.

Dalpanagioti Thomai, 2018a, “A Frame-semantic Approach to Co-occurrence Patterns: A Lexicographic Study of English and Greek Motion Verbs”, International Journal of Lexicography 31(4), 420-451.

Dalpanagioti Thomai, 2018b, “Corpus-based cognitive lexicography: Insights into the meaning and use of the verb stagger”, in Čibej Jaka, Gorjanc Vojko, Kosem Iztok & Krek Simon (Eds.), Proceedings of the XVIII EURALEX International Congress: Lexicography in Global Contexts, Ljubljana: University of Ljubljana, 649-662.

Dalpanagioti Thomai, 2019, “From corpus usages to cognitively informed dictionary senses: Reconstructing an MLD entry for the verb float”, Lexicography: Journal of ASIALEX 6(2), 75-104.

Ferraresi Adriano, Zanchetta Eros, Bernardini Silvia & Baroni Marco, 2008, “Introducing and evaluating ukWaC, a very large web-derived corpus of English”, in Evert Stefan, Kilgarriff Adam & Sharoff Serge (Eds.), Proceedings of the 4th Web as Corpus Workshop (WAC-4) – Can we beat Google?, http://wacky.sslmit.unibo.it/lib/exe/fetch.php?media=papers:wac4_2008.pdf.

Fillmore Charles, 1982, “Frame semantics”, Linguistics in the Morning Calm, Seoul: Hanshin Publishing, 111–137. Reprinted in Geeraerts Dirk (Ed.), 2006, Cognitive Linguistics: Basic Readings, Berlin: Mouton de Gruyter, 373-400.

Fillmore Charles, 1985. “Frames and the semantics of understanding”, Quaderni di Semantica 6(2), 222-254.

Fillmore Charles & Baker Collin, 2010, “A frames approach to semantic analysis”, in Heine Bernd & Narrog Heiko (Eds.), The Oxford Handbook of Linguistic Analysis, Oxford: Oxford University Press, 313-339.

Fillmore Charles & Petruck Miriam, 2003, “FrameNet glossary”, International Journal of Lexicography 16(3), 359-361.

FrameNet, https://framenet.icsi.berkeley.edu/fndrupal.

Gantar Polona, Colman Lut, Parra Escartín Carla & Martínez Alonso Héctor, 2019, “Multiword expressions: Between lexicography and NLP”, International Journal of Lexicography 32(2), 138-162.

Goddard Cliff, Wierzbicka Anna & Wong Jock, 2016, “‘Walking’ and ‘running’ in English and German: The conceptual semantics of verbs of human locomotion”, Review of Cognitive Linguistics 14(2), 303-336.

Hanks Patrick, 2004, “Corpus Pattern Analysis”, in Williams Geoffrey & Vessier Sandra (Eds.), Proceedings of the XI EURALEX International Congress, Lorient: UBS, 87-98.

Hanks Patrick, 2013a, Lexical Analysis: Norms and Exploitations, Cambridge, MA: MIT Press.

Hanks Patrick, 2013b, “English and American II: Synchronic lexicography”, in Gouws R. H., Heid Ulrich, Schweickard Wolfgang & Wiegand Herbert Ernst (Eds.), Dictionaries. An International Encyclopedia of Lexicography. Supplementary Volume: Recent Developments with Focus on Electronic and Computational Lexicography, Berlin: Mouton de Gruyter, 720-730.

Hanks Patrick, 2020, “How context determines meaning”, in Corpas Pastor Gloria & Colson Jean-Pierre (Eds.), Computational Phraseology, Amsterdam/Philadelphia: John Benjamins, 297-309.

Hanks Patrick, 2021, Corpus Pattern Analysis. CPA Project Page, http://nlp.fi.muni.cz/projects/cpa.

Hanks Patrick & Franklin Emma, 2019, “Do online resources give satisfactory answers to questions about meaning and phraseology?”, in Corpas Pastor Gloria & Mitkov Ruslan (Eds.), Computational and Corpus-Based Phraseology, New York: Springer, 159-172.

Jarosz Izabela, 2019, “Verb-particle constructions in Cognitive Linguistics perspective: Compositionality behind selected English phrasal verbs”, Linguistics Beyond and Within (LingBaW) 5(1), 29-45.

Kövecses Zoltán, 2005, Metaphor in Culture. Universality and Variation, Cambridge: Cambridge University Press.

Kövecses Zoltán, 2021, “Standard and extended conceptual metaphor theory”, in Wen Xu & Taylor John (Eds.), The Routledge Handbook of Cognitive Linguistics, New York/London: Routledge, 191-203.

Kövecses Zoltán & Csábi Szilvia, 2014, “Lexicography and cognitive linguistics”, Revista Española de Lingüística Aplicada/Spanish Journal of Applied Linguistics 27(1), 118-139.

Lakoff George & Johnson Mark, 1980, Metaphors We Live By, Chicago/London: The University of Chicago Press.

Lew Robert, 2013, “Identifying, ordering and defining senses”, in Jackson Howard (Ed.), The Bloomsbury Companion to Lexicography, London: Bloomsbury, 284-302.

Mahpeykar Nanges & Tyler Andrea, 2014, “A principled cognitive linguistics account of English phrasal verbs with up and out”, Language and Cognition 7(1), 1-35.

Meer Geart van der, 1999, “Metaphors and dictionaries: The morass of meaning, or how to get two ideas for one”, International Journal of Lexicography 12(3), 195-208.

Moon Rosamund, 1987, “The analysis of meaning”, in Sinclair John (Ed.), Looking Up: An Account of the COBUILD Project in Lexical Computing, London: Collins, 86-103.

Moon Rosamund, 2004, “On specifying metaphor: An idea and its implementation”, International Journal of Lexicography 17(2), 195-222.

Perdek, Magdalena, 2010, “Getting through to phrasal verbs: A cognitive organization of phrasal verb entries in monolingual pedagogical dictionaries of English”, in Dykstra Anne & Schoonheim Tanneke (Eds.), Proceedings of the XIV Euralex International Congress, Leeuwarden: Fryske Akademy, 1390-1398.

Rodríguez-Puente, Paula, 2019, The English Phrasal Verb, 1650-Present: History, Stylistic Drifts, and Lexicalization, Cambridge: Cambridge University Press.

Ruiz de Mendoza Francisco, 2021, “Conceptual metonymy theory revisited. Some definitional and taxonomic issues”, in Wen Xu & Taylor John (Eds.), The Routledge Handbook of Cognitive Linguistics, New York/London: Routledge, 204-227.

Ruiz de Mendoza Francisco & Galera-Masegosa Alicia, 2011, “Going beyond metaphtonymy: Metaphoric and metonymic complexes in phrasal verb interpretation”, Language Value 3(1), 1-29.

Ruppenhofer Josef, Ellsworth Michael, Petruck Miriam, Johnson Christopher & Scheffczyk Jan, 2016, FrameNet II: Extended Theory and Practice, https://framenet2.icsi.berkeley.edu/docs/r1.7/book.pdf.

Sinclair John, 1987, Looking Up: An Account of the COBUILD Project in Lexical Computing, London: Collins.

Talmy Leonard, 1985, “Lexicalization patterns: semantic structure in lexical forms”, in Shopen Timothy (Ed.), Language Typology and Syntactic Description 3: Grammatical Categories and the Lexicon, Cambridge: Cambridge University Press, 57-149.

Talmy Leonard, 2000, Toward a Cognitive Semantics, Cambridge, MA: MIT Press.

Taylor John, 1993, “Some pedagogical implications of cognitive linguistics”, in Geiger Richard & Rudzka‐Ostyn Brygida (Eds.), Conceptualizations and Mental Processing in Language, Berlin: Mouton de Gruyter, 201-223.

Thim Stefan, 2012, Phrasal Verbs: The English Verb–Particle Construction and its History, Berlin/New York: Mouton De Gruyter.

Torres-Martínez Sergio, 2018, “Applied Cognitive Construction Grammar: A usage-based approach to the teaching of phrasal verbs (and other constructions)”, European Journal of Applied Linguistics 6(2), 279-314.

Wahl Alexander & Gries Stefan, 2020, “Computational extraction of formulaic sequences from corpora: Two case studies of a new extraction algorithm”, in Corpas Pastor Gloria & Colson Jean-Pierre (Eds.), Computational Phraseology, Amsterdam/Philadelphia: John Benjamins, 83-110.

Zhi Lu & Juan Sun, 2015, “A view of research on English polysemous phrasal verbs”, Journal of Literature and Art Studies 5(8), 649-659.

Dictionaries

CALD: Cambridge Advanced Learner’s Dictionary, http://www.dictionary.cambridge.org.

COBUILD: COBUILD Advanced English Dictionary, http://www.collinsdictionary.com.

LDOCE: Longman Dictionary of Contemporary English, http://www.ldoceonline.com.

MEDAL: Macmillan English Dictionary for Advanced Learners, http://www.macmillandictionary.com.

OALD: Oxford Advanced Learner’s Dictionary, http://www.oxfordlearnersdictionaries.com.

PDEV: Pattern Dictionary of English Verbs, http://www.pdev.org.uk/#browse?q=;f=C.

Top of page

Notes

1 In this study we use the term “phrasal verb” following Thim’s [2012: 4] note that this term usually appears in discussions of “the Modern English verb-particle construction and its peculiarities” while the term “particle verb” or “verb-particle construction” is more neutral and is used in comparative studies.

2 CPA analysis of PVs like blow up and wipe out is further illustrated in Hanks [2013b: 729; 2020: 303] and Hanks & Franklin [2019: 160].

3 Core FEs are the conceptually necessary components of a frame; indications of time, place and manner are typically non-core, i.e. peripheral, since they do not uniquely characterize a frame (Fillmore & Petruck [2003: 359]).

4 FrameNet uses font and colour differences for the names of frames and FEs. The present work uses small capitals for the names of FEs and square brackets for frames.

5 For information on the other types of null instantiation that FrameNet recognizes but do not appear in Figure 3, i.e. INI (Indefinite Null Instantiation) and CNI (Constructional Null Instantiation), see Atkins, Fillmore & Johnson [2003: 268-270].

6 For example, Ruiz de Mendoza & Galera [2011: 20] analyze the PV burst into (as in “He burst into tears.”) based on a metonymy (effect for cause) built into the target domain of a double source metaphoric amalgam, which integrates the metaphors emotional damage is physical damage and emotional damage is motion. For an example of a metaphoric chain interacting with metonymy, see the analysis of the PV fed up with (as in “Eventually someone got fed up with her behavior and called the cops.”) (Ruiz de Mendoza & Galera [2011: 23]).

7 This semantic order approach, which argues that the core meaning should precede extended uses, stands in contrast to the frequency-based approach, according to which highly frequent metaphorical extensions should precede less frequently occurring literal meanings (Hanks [1987: 133-134]).

8 For a similar discussion of the “verb + particle” combination as a compositional construction and as a phrasal verb “where, according to one widely accepted criterion, a meaning cannot be assigned separately to the individual components”, see Hanks [2020: 304].

9 Frame definition: “A competitor claims a prize as a result of the outcome of their participation in a competition.” (FrameNet)

10 Frame definition: “These are words describing situations in which a perpetrator takes goods from a victim or a source.” (FrameNet)

11 Frame definition: “A controlling_entity, controlling_situation, or controlling_variable control a dependent_entity, dependent_situation, or dependent_variable. The latter, dependent, element or some aspect of it is not just influenced, but determined by the controlling element.” (FrameNet)

12 Only one relevant instance has been found in the BNC with off instead of away: “It was heartfelt and deep, but Theo should not run off with the idea that it was all laughter and light and the cooing of turtle doves”.

13 Frame definition: “The self_mover, a living being, moves under its own direction along a path. Alternatively or in addition to path, an area, direction, source, or goal for the movement may be mentioned.” (FrameNet)

14 Frame definition: “An object (the theme) moves away from a source. The source may be expressed or it may be understood from context, but its existence is always implied by the departing word itself.” (FrameNet)

15 Frame definition: “A recipient starts off without the theme in their possession, and then comes to possess it. Although the source from which the theme came is logically necessary, the recipient and its changing relationship to the theme is profiled.” (FrameNet)

16 For an analysis of the self metaphors, see Kövecses [2005: 55-56].

17 It is thus clear why we treat the pattern [ideas/ feelings] run away with [personal pronoun] under the construction walk/run away/off with something (as opposed to somebody); it might seem that the with- slot is filled with an animate entity (somebody) but essentially this is conceived as an object (something).

Top of page

List of illustrations

Title Figure 1. CPA patterns taken from the PDEV entry for blow
URL http://journals.openedition.org/lexis/docannexe/image/6186/img-1.png
File image/png, 160k
Title Figure 2. Part of FrameNet’s [Destroying] frame report
URL http://journals.openedition.org/lexis/docannexe/image/6186/img-2.png
File image/png, 232k
Title Table 5. Accounting for the meaning and use of walk/run away/off with something
URL http://journals.openedition.org/lexis/docannexe/image/6186/img-3.png
File image/png, 99k
Title Figure 4. A reconstructed entry for the PV construction
URL http://journals.openedition.org/lexis/docannexe/image/6186/img-4.png
File image/png, 211k
Top of page

References

Electronic reference

Thomai Dalpanagioti, Motivated patterns of phrasal verbs and learner’s dictionariesLexis [Online], 19 | 2022, Online since 26 March 2022, connection on 24 May 2022. URL: http://journals.openedition.org/lexis/6186; DOI: https://doi.org/10.4000/lexis.6186

Top of page

About the author

Thomai Dalpanagioti

Aristotle University of Thessaloniki, Greece
thomdalp@enl.auth.gr

By this author

Top of page
Search OpenEdition Search

You will be redirected to OpenEdition Search