Metonymy in Language about Organizations: A Corpus-Based Study of Company Names

Metonymy in Language about Organizations:A Corpus-Based Study of Company Names

Joep P. CornelissenLeeds University Business School, University of Leeds

abstract In this paper, I examine the use of metonymies in people’s talk aboutorganizations. Drawing upon a corpus of natural talk extracted from the British NationalCorpus (BNC) I identify recurring categories of metonymies that appear to be a central partof people’s talk about organizations. These categories of metonymies involve substitutionswhere an organization stands in for its members, its products, its facilities, its stock or shares ora company-related event. I also found that metonymies in each of these categories are used asbasic metonymic expressions that are only partially connected to metaphorical expressions andinterpretations of organizations. Where those connections exist, the use of metonymies followsa metaphor-from-metonymy linguistic pattern (where a metaphorical meaning arises from theuse of a metonymy) rather than a metonymy-within-metaphor pattern (where a metonymy ispart of a metaphorical expression). I elaborate on the implications of these findings for ourunderstanding of how organizations are discursively constructed and understood throughmetonymic language.

INTRODUCTION

How do people talk about and understand organizations? In their talk about organiza-tions, scholars and practitioners alike frequently use creative and figurative forms oflanguage to produce new, coherent representations of organizations and organizationallife. There has been a longstanding interest in one particular form of language regardingorganizations: how scholars and managers use and interpret metaphorical word combi-nations or expressions involving an adjective and a noun (e.g. organizational identity) ortwo nouns adjacent to one another or in an extended phrase (e.g. business strategy) (e.g.Alvesson, 1993; Astley and Zammuto, 1992; Cornelissen, 2005; Mauws and Phillips,1995; Morgan, 1980, 1983; Putnam et al., 1996; Tsoukas, 1991). Besides such meta-phorical word combinations or expressions, there are however also other forms oflanguage including noun-noun compounds (e.g. ‘organization studies’) and the tropes of

Editors’ Note: This paper was under review with JMS prior to Joep Cornelissen’s appointment as AssociateEditor of JMS.

Address for reprints: Joep P. Cornelissen, Leeds University Business School, Maurice Keyworth Building,University of Leeds, Leeds LS2 9JT, UK ([email protected]).

© Blackwell Publishing Ltd 2007. Published by Blackwell Publishing, 9600 Garsington Road, Oxford, OX4 2DQ, UKand 350 Main Street, Malden, MA 02148, USA.

Journal of Management Studies 45:1 January 2008doi: 10.1111/j.1467-6486.2007.00737.x

mailto:[email protected]

metonymy, synecdoche and irony that are part of the generative lexicon of organiza-tional scholars and individuals working in organizations; i.e. language constructions thatare used not only to reference and communicate ideas but also to generate understand-ings of organizations (Poole and Van de Ven, 1989).

Because of the traditionally strong focus on metaphors, the role of other tropes,particularly metonymy (a figure of speech that involves a part-whole or whole-partsubstitution), within the language about organizations has been largely ignored (Oswicket al., 2003; Oswick et al., 2004; Putnam and Fairhurst, 2001). Commentaries andempirical research on metonymy to date (Manning, 1979; Musson and Tietze, 2004;Putnam, 2004; Watson, 1995) have been restricted to local and specific uses of metony-mies within small samples of managerial or employee discourse. Although insightful,these studies say very little about how often and what kinds of metonymies feature across

people’s talk about organizations when we account for different genres such as academicand professional as well as lay talk about the subject of organizations. I therefore presenta large-scale study of the use of metonymy in talk about organizations extracted from theBritish National Corpus (BNC). The British National Corpus (BNC) is a 100 millionword collection (‘corpus’) of samples of written and spoken language from a wide rangeof sources, designed to represent a wide cross-section of British English from the laterpart of the 20th century, both spoken and written.[1]

With this study, my aims are: (1) to identify the incidence of metonymies in talk aboutorganizations; (2) to identify the different kinds or categories of metonymies that peopleuse when talking about organizations; and (3) to analyse whether and how metonymiesare connected to metaphorical expressions concerning organizations. The latter aimfollows from work in organization studies that has suggested that metonymies intimateand depend on metaphorical images of organization (e.g. Hamilton, 1997; Manning,1979; Morgan, 1983, 1996; Musson and Tietze, 2004; Oswick and Grant, 1996; Oswicket al., 2003). That is, metonymies are seen to be a part of, and indeed motivated by, alarger metaphorical expression that features in people’s talk. Goossens (1990, 1995a,1995b) labelled this use of metonymy as metonymy-within-metaphor. Besides this lin-guistic form, Goossens (1990, 1995a, 1995b) also pointed to alternative combinationsbetween the metonymic and the metaphoric including a prominent metaphor-from-metonym form, where the metaphorical meaning arises, or follows from, the use of ametonymy. The large-scale nature of the corpus of talk extracted from the BNC allowsme to answer these questions about the incidence and kinds of organizational metony-mies in people’s talk and about the actual connections between the metonymic and themetaphoric (Deignan, 2005). In doing so, the results of the study contribute to a greaterunderstanding not only of how people use metonymy to discursively frame or refer toorganizations and reason about them (Gibbs, 1994; Lakoff, 1987; Lakoff and Johnson,1980, 1999) but also of how the metonymic connects with the metaphoric in ourdiscursive constructions of what organizations are (Cooren, 2004; Cornelissen, 2005;Taylor and Cooren, 1997; Taylor and Van Every, 2000). The study is of relevance toscholars in management and organization studies with an interest in understanding howpeople talk about and understand organizations as entities that have specific roles withinthe economy and society at large. As I will demonstrate, the way in which people do thisinvolves to a significant extent metonymic and metaphoric expressions.

J. P. Cornelissen80

© Blackwell Publishing Ltd 2007

METONYMY (AND METAPHOR) IN LANGUAGE ABOUTORGANIZATIONS

A large body of research in organization studies emphasizes the fundamental role oflanguage in representing and understanding organizations (e.g. Alvesson and Kärreman,2000; Astley and Zammuto, 1992). Both organizational scholars and people withinorganizations rely upon language to identify and represent organizations, or aspects ofthem, in order to grasp and understand them – given that organizations cannot be simplyrepresented, experienced or understood as single objects or entities (Cooren et al., 2005;Sandelands and Srivatsan, 1993; Weick, 1989). Weick and Daft argued in this respectthat organizations are ‘vast, fragmented, elusive, and multidimensional’ (1983, p. 72) andcan be talked about and understood as, for example, ‘input-output systems, resourceallocation systems, collections of humans with needs to be met, growth and survivalsystems, tools in the hands of goal-setters, coalitions of interest groups [and] transfor-mation systems’ (1983, p. 172). Their suggestion was that although we can unpack theconcept of organization and talk about it in literal terms as involving, for example, acollective of people who work together, buildings, material resources, etc, we often domore than that. That is, in the language that we use as academics as well as in ourday-to-day talk as people experiencing and interacting with organizations we talk orga-nizations ‘into existence’ through language uses such as metaphor, metonymy and irony(Cooren, 2004; Cooren et al., 2005; Sandelands and Srivatsan, 1993; Taylor and VanEvery, 2000). These language uses allow us to transfer meaning and inferences to theconcept of organizations and in doing so enhance our abilities to reason about them.

Many aspects of our experience cannot be clearly delineated in terms of the naturallyemergent dimensions of our experience. This is typically the case for human emotions,abstract concepts, mental activity . . . Though most of these can be experienceddirectly, none of them can be fully comprehended on their own terms. Instead, wemust understand them in terms of other entities and experiences, typically other kindsof entities and experiences. (Lakoff and Johnson, 1980, p. 177)

The classic tropes of metaphor, metonymy, synecdoche and irony (Burke, 1969;Brown, 1977; Manning, 1979), therefore, are important forms of language use not onlyto represent organizations but also to generate meanings and understandings about them. Inother words, these tropes are not ‘empty’ rhetorical figures of speech or a playful use ofwords, but are important ways in which people talk about organizations (Taylor and VanEvery, 2000; Weick, 2006); that is, give meaning to organizations and understand themin a particular way.

A metaphor can be defined as a use of language in which the one term or concept(called the target) is compared to another term (called the source), with the sourcestemming from a domain of knowledge and language use that is not typically associatedwith the target (e.g. Cornelissen, 2005, 2006; Morgan, 1983; Oswick et al., 2002;Tsoukas, 1991). For example, when we compare an organization to a complex adaptiveor chaotic system (e.g. Eisenhardt and Bhatia, 2002), we are mobilizing terminology (andknowledge) from biophysics to describe the nature of organizations. For this reason,metaphor researchers often refer to a semantic clash or anomaly between the target and

Metonymy in Language about Organizations 81


source when they are first compared (Cornelissen, 2005). Unlike metaphors whichinvolve a comparison between two concepts or terms from domains that are (at leastinitially) seen as distant from one another, metonymy and synecdoche both rely upon anexchange between parts within the same domain of language use and knowledge.Metonymy and synecdoche involve a contiguous mapping and, typically, involve apart-whole or whole-part substitution. Metonymy is sometimes defined as a part-wholesubstitution, whereas synecdoche is seen as the reverse: the whole is employed torepresent the part. As Manning (1979, p. 662) writes:

Metonymy takes the whole (an organization) to be indicated by its parts (e.g., thenumber of levels in an organization, the size of the body of rules governing procedures,the rates of mobility between and within organizational slots). The whole is thusrepresented by the parts; the essential features of a whole are reduced to indices.

In the present study, I follow writings in linguistics where metonymy is a figure ofspeech, in which one expression is used to refer to the standard referent of a related one(Gibbs, 1994; Lakoff, 1987; Lakoff and Johnson, 1980, 1999) and may involve both apart-whole or whole-part substitution. For example, in the idiomatic expression ‘themanager has his eye on the company targets’ the manager’s eye stands in for his/herentire ability to see and by implication for his determination to orient himself towards thecompany’s targets. Similarly, in order to imbue an expression with gravitas, a managermight say to a subordinate ‘the company is unhappy with your work performance’. Inthis instance, the whole – the company – is invoked as the equivalent of the part – themanager, when the manager is in fact unhappy with the subordinate’s work (Grant andOswick, 1996; Oswick et al., 2003). Irony, like metaphor, entails mappings acrossdomains, but is based upon the ‘juxtaposing of opposites’ (Oswick et al., 2003; Poole andVan de Ven, 1989). It involves the use of the inappropriate in order to describe some-thing in a paradoxical and contradictory way. An example is talking of spaghetti oranarchy as a good form of organization (e.g. Foss, 2003; Morgan, 1983).

For the most part, research on organizational metaphors overshadows studies onmetonymy (or synecdoche) and irony (Oswick et al., 2003, 2004; Putnam and Fairhurst,2001). The limited research that exists on metonymy has examined how within mana-gerial or employee talk whole-part relationships develop alternative meanings and newpatterns of association. In a study of police discourse, Manning (1979) illustrated how theconcept of drug use becomes a crime through aligning this whole with parts of thecriminal process, such as crime statistics, seizure data, and number of warrants. As heobserved, ‘indicators of larger processes [of drug use] were shrunken into microformatsand [were] modeled using crime statistics, seizure data, number of warrants served andthe like’ (Manning, 1979, p. 666). Putnam (1995, 2004) demonstrated how the termslanguage and money in a teachers’ bargaining moves from referencing sections of thecontract to signifying competing commodities that serve as a formula for a settlement.Musson and Tietze (2004) focused on organizational names as metonymic of physicalplaces and spaces within the talk of faculty members of a university. Cooren and hiscolleagues (Cooren, 2004; Cooren et al., 2005, 2006) have demonstrated how when weattempt to understand organizations we instantiate them in many concrete ways as

J. P. Cornelissen82


collections of concrete ‘things’ such as texts, buildings, machines, people and materials.Such instantiations or ‘embodiments’, which are situated in a particular context and timeframe and happen through metonymic contiguity mappings, allow as Cooren et al.(2005) suggest for an ‘emergent’ (rather than reductive or reified) and ‘plurified’ under-standing of what an organization is. Metonymy also entered into Watson’s (1995) studyof two managers in a telecommunications company deliberating about a personal devel-opment program within workplace change. Thus, in these studies, members of organi-zations use metonymy in everyday talk not only to name and refer to organizations andorganizational life (through reference shifts in language), but also symbolically to legiti-mate their activities, reaffirm status relations, and convince themselves about interpre-tations of an organizational event.

Although these studies have started to provide rich insights into the role of metonymyin managerial or employee talk, when taken together they point to an important limi-tation in our current understanding of metonymies in language within or about organi-zations. That is, the empirical data on which these studies are based involves local andsmall episodes of talk (between members of an organization during a reorganization,organizational change, negotiations etc), rather than more broad based corpora oflanguage or text analyzed for metonymies. As a result, our collective understanding ofmetonymy across organizational talk is still rather limited. This is the case, despite thesuggestion coming from linguistics that many metonymies are actually quite regularbeyond specific episodes of talk. That is, linguistic studies (Fass, 1997; Lakoff andJohnson, 1980; Nissim and Markert, 2005) have postulated conventionalized metonymicpatterns (e.g. organization-for-product) that operate on semantic base classes such as‘organization’.

At this point it is important to draw a distinction between discourse analysisapproaches (Phillips and Hardy, 2002) that consider the use and meaning of metonymiesas socially constructed and linguistic approaches that consider linguistic usage as rela-tively more ‘fixed’ across different speakers and social contexts. In other words, alinguistic approach assumes that speakers of a language share particular conventionalmodes of talking, including conventional ways of using metonymies when referring toorganizations. The last few years have seen a rise in the use of electronic corpora forlinguistic research on people’s talk (e.g. Deignan, 2005). The use of such corpus data doesnot, of course, necessarily lead to findings that could not have been arrived at via othermethods. However, corpus data enable researchers to study naturally occurring linguisticpatterns on a large scale and therefore tend to provide a greater number and variety ofexamples than can be generated by small amounts of data. The advantage of a corpus-based linguistic analysis is that it allows the researcher to place a claim about the use ofparticular metonymies in talk about organizations on a much broader and explicitempirical footing because it is based on a large collection of natural talk rather than asmall set of interview or speech transcripts (Deignan, 2005).

In this study, I demonstrate the use of a corpus-based linguistic analysis of metonymiesin a corpus of talk involving company names. The identification of metonymies in thecorpus is focused on company names as company names are a direct embodiment oforganizations in naturally occurring talk (Cooren et al., 2005). Whereas prior researchhas largely focused on small episodes of talk within an organizational context; this corpus



allows me to address the use of organizational metonymies directly by focusing onmetonymies involving such company names. The rationale behind this is that adocumentation of how people use metonymic whole-part mappings through the use ofthese company names provides fundamental insights into how they frame and refer toorganizations more generally.

Connections between Metonymy and Metaphor

The systematic and comprehensive nature of this corpus-based analysis of organizationalmetonymies also allows me to examine the well-established claim that metonymies aregrounded in, and indeed depend on, metaphorical images of organization (e.g. Hamilton,1997; Manning, 1979; Morgan, 1983, 1996; Musson and Tietze, 2004; Oswick andGrant, 1996; Oswick et al., 2003). Morgan argued in his early writings that metaphor‘makes meaning in a primal way’ (1983, p. 602), i.e. that it is the source of comparisonbetween two different objects or situations; and that metonymy, synecdoche and irony areportrayed as ‘secondary forms within the domain or context forged through metaphor’(Morgan, 1983, p. 602). More recently, Morgan acknowledged a mutual interdependencebetween the metaphorical and metonymical; ‘metaphor and metonymy are alwaysinterconnected. You cannot have one without the other’ (Morgan, 1996, p. 231). That is,while he argues that ‘a metaphorical image relies on some kind of metonymical reduction,otherwise it remains thin air’ (Morgan, 1996, p. 231), Morgan equally suggests that‘metonymy is entirely dependent on metaphor, for without a prefiguring image we havenothing to see’ (Morgan, 1996, p. 231). Morgan’s (1983, 1996) views regarding theintimate and necessary connection of the metaphorical and metonymical resonates withwork in the linguistics community. Jakobson (1990), for example, has argued thatmetaphors and metonymies provide the ‘bipolar structure’ of language; both are neces-sary for the development of discourse and meaning through statements of similarity(metaphor) and contiguity (metonymy) and both mutually implicate each other in actualinstances of language use. Following on from Jakobson, a number of linguists have arguedthat the predominant connection between metonymy and metaphor is one wheremetonymies rely upon metaphors and are therefore seen as a subclass of metaphor (e.g.Lakoff and Johnson, 1980; Levin, 1991; Searle, 1979). Goossens (1990, 1995a, 1995b)labelled this connection between a metonymy and metaphor as ‘metonymy-within-metaphor’, which occurs when ‘a metonymically used entity is embedded within a(complex) metaphorical expression’ (Goossens, 1995a, p. 172). One of the examples thatGoossens (1995a) gives is the expression ‘bite one’s tongue off ’ where a tongue is usedmetonymically to stand for speech and where the expression as a whole is used meta-phorically to mean ‘deprive oneself of the facility of speech’. As with this example,metonymy-within-metaphor can only describe multi-word expressions, as a single wordcould not usually be said to contain a metonymy within a larger expression or unit ofmeaning. Goossens (1995a, 1995b) also pointed to another category, ‘metaphor-from-metonymy’ which has been found to be a very frequent kind of ‘connection’ or ‘interac-tion’ between metonymy and metaphor (Deignan, 2005; Deignan and Potter, 2004;Goossens, 1990, 1995a, 1995b). In this category, an expression develops a meaningthrough metonymy, a meaning that is then mapped metaphorically onto another domain

J. P. Cornelissen84


or cues a further metaphorical interpretation. One example would be the abovemen-tioned example of ‘the manager has his eye on the company targets’ which followingGoossens (1995a, 1995b) can be understood in two different ways; firstly it can mean‘viewing’ through metonymy with the manager’s eye standing for the act of seeing.Alternatively, it may refer to an imagined scene where a manager is physically focusingon, and moving closer to, the company’s targets. In this case, the expression is a metaphor,but one which is derived through a metonymy. The main difference between the twokinds of connections is that in ‘metonymy-within-metaphor’ one word of the largerexpression is a metonymy whereas the interpretation of the larger expression is meta-phorical. In ‘metaphor-from-metonymy’ the whole expression is initially a metonymy,but then gives rise to a further metaphorical interpretation. Distinguishing the two, then,may involve deciding whether the whole expression is metonymically-based, or if just oneelement of it is (Deignan, 2005).

Goossens (1990, 1995a, 1995b) also identified two further connections between themetonymic and the metaphoric (labelled as ‘metaphor-within-metonymy’ where a meta-phor is embedded in a metonymy and tends to ‘metaphorize’ the whole expression and‘demetonymization in a metaphorical context’ where a metonymic reading is relevantbut abandoned in the context of a metaphorical interpretation), but both his ownresearch and that of other linguists (Deignan, 2005; Deignan and Potter, 2004) hasestablished that these are extremely rare in actual instances of language use.

METHOD

In order to answer the research questions (What is the incidence of metonymies in talkabout organizations? What different kinds or categories of metonymies feature in peo-ple’s talk about organizations? To what extent and how are metonymies connected tometaphorical expressions concerning organizations?), I needed a large enough corpus oftalk about organizations to identify the range of metonymies that people use. I thereforedecided to use Nissim and Markert’s (2005) database of metonymies in talk aboutorganizations, and to reanalyse their data in line with the three research questions. Thesampling frame for the Nissim and Markert database consisted of the Fortune 500company names (http://www.fortune.com/fortune/fortune500),[2] plus alternativespellings, acronyms, and abbreviations, for a total of 528 different company names.Nissim and Markert (2005) randomly extracted 3100 instances of these names from theBritish National Corpus (BNC), allowing any name in the sampling frame to occur. Allsamples include three sentences of context. The BNC is a 100 million word collectionof written and spoken language from a wide range of sources, designed to represent awide cross-section of current British English, both spoken and written (see http://www.natcorp.ox.ac.uk/). The BNC includes, for example, extracts from regional andnational newspapers, specialist periodicals and journals for all ages and interests, aca-demic books and popular fiction, published and unpublished letters and memoranda,school and university essays, among many other kinds of text. As such, the BNCincorporates academic and professional as well as lay talk about the subject of organi-zations. One of the advantages, therefore, of using the BNC is that this corpus cuts acrossgenres of spoken and written discourse, and is as such representative of a range of



http://www.fortune.com/fortune/fortune500

http://www.natcorp.ox.ac.uk

http://www.natcorp.ox.ac.uk

different ways in which people (in spoken and written discourse) talk about organizations.The BNC is also large enough to be used for general linguistic research, as opposed tosmaller corpora or collections of texts (e.g. interviews with managers) that often onlyallow for the analysis of a particular genre (e.g. managerial talk) and where generalizationabout general language use is often not possible. The use of the BNC allows me to makeobservations about the frequency and idiomatic nature of particular metonymies, andtheir connections with metaphorical expressions, which would not have been possiblewith a smaller corpus or collection of texts. A further advantage of using a large corpusis that any one speaker (or researcher) will not know all the words of their language andtheir meaning in use: in my analysis of organizational metonymies, for example, I foundthat metonymies that follow a metaphor-from-metonymy linguistic pattern were muchmore frequently used than metonymies following a metonymy-within-metaphor pattern,which was surprising given the strong and continuing emphasis in organization studieson metonymies within metaphors (e.g. Morgan, 1983, 1996, 2006; Oswick and Grant,1996; Oswick et al., 2003). Corpus data can thus help to provide a more systematic andinformed analysis of language. Linguists such as Sinclair (1991) have argued in thisrespect that the systematic study of corpus data yields information about language usethat is otherwise not available:

. . . the contrast exposed between the impressions of language detail noted by people,and the evidence compiled objectively from texts is huge and systematic. It leads oneto suppose that human intuition about language is highly specific, and not at all a goodguide to what actually happens when the same people actually use the language.(Sinclair, 1991, p. 4)

Using Nissim and Markert’s (2005) database, my analysis focuses on company namesas these involve expressions in which people talk about organizations. Talk is heredefined as people’s (including professional managers, journalists at news media andmembers of the general public) talk about organizations in natural settings such as innews articles or letters, which is distinguished from talk uttered in research settings andthat may be influenced and/or controlled by a researcher.

For the analysis of the data, I followed cognitive and corpus linguistic studiesof metonymy which provide strong evidence that metonymic reference shifts (e.g.organization-for-product) occur systematically for a certain base class such as organiza-tion (e.g. Lakoff, 1987; Lakoff and Johnson, 1980, 1999). Guided by this assumption, I setout to identify general categories of metonymies in the data, following a particular codingscheme. This coding scheme, as explained below, distinguishes between literal readingsinvolving company names and a set of metonymic patterns for the base class of companynames. These metonymic patterns were derived from observations made by Nissim andMarkert (2005) about patterns in metonymy use about organizations.

Coding and categorizing the data in this way, I am making inferences about patternsin the use of metonymies across different contexts of language use, and in doing sointerpret their use and meaning across these contexts. My purpose here is to identify andinterpret general linguistic patterns of metonymy use, rather than establishing the fullrange of potential and contextualized meanings of any particular metonymic expression.

J. P. Cornelissen86


The coding scheme followed written guidelines, whose main features are replacementtests (e.g. if an occurrence of ‘BP’ can be replaced by ‘shares of BP’, it was coded asorganization-for-index), examples for each category and instructions for ambiguouscases. The general coding scheme, derived from Nissim and Markert (2005), distin-guishes as mentioned between literal, metonymic and mixed readings for a base class; inthis case company names.[3] The literal reading refers to the organization as a legal entitywhich has a charter or defined aims (cf. Czarniawska-Joerges, 1994). This includesdescriptions of the organization’s structure (see (1)), and relations between organizationsand the products/services that they offer (see (2)).

(1) NATO members [. . . ]

(2) Intel’s Indeo video compression hardware [. . . ]

Metonymic readings cover metonymies that follow regular patterns as well as metony-mies that do not. Most patterns are specific to the base class of company names. Theorganization-specific patterns are distinguished in Table I.

These patterns form the categories that I used for the coding of the entire corpus.Examples 3–8 illustrate these categories, as derived from the initial 400 samples, whichincluded 125 instances of company names.

(3) Last February NASA announced [. . . ]

(4) It’s customary to go to work in black or white suits. [. . . ] Woolworths wear them

(5) And he hasn’t bought me a Renault [. . . ]

(6) The opening of a McDonald’s is a major event

(7) Eurotunnel was the most active stock

(8) [. . . ] the resignation of Leon Brittan from Trade and Industry in the aftermath of

Westland[4]

Besides these conventional class-specific patterns, two class-independent metonymic pat-terns can be applied to most nouns. In object-for-name shifts, a word is used as a meresignifier rather than referentially, as Chevrolet and Ford in (9).

(9) Chevrolet is feminine because of its sound (it’s a longer word than Ford, has an open vowel

at the end, connotes Frenchness)

Table I. Organization-specific metonymic patterns

org-for ⟨org⟩ stands for Examples

Members an official who acts for ⟨org⟩, or all members of ⟨org⟩ (3), (4)Product the product(s) an ⟨org⟩ produces (5)Facility the facility that houses ⟨org⟩ or one of its branches (6)Index an index, like a stock index, indicating the value of ⟨org⟩ (7)Event an event associated with ⟨org⟩, e.g., a scandal (8)



In object-for-representation metonymies, a name refers to a representation (such as aphoto or painting) of the standard referent. The logo of an organization, for example, canbe regarded as its representation, as in (10).

(10) BT’s pipes-of-Pan motif was, for him, somehow too British. Graphically, it lacked

what King calls the ‘world class’ of IBM, Apple Computer, Ford, Sony, and

Shell.

In some examples, two predicates trigger a different reading each, thus yielding amixed reading (cf. Nunberg’s (1995) account of co-predication). Example 11 evokes anorg-for-index as well as an org-for-members reading (triggered by ‘slipped’ and ‘con-firming’, respectively).

(11) Barclays slipped 4p to 351p after confirming 3,000 more job losses.

In addition, following Nissim and Markert (2005) I also distinguish a category ‘other’which covers unconventional metonymies (as in, for example, ‘Funds for OperationShakespeare had been paid into Barclays Bank, he said, and when the mission was overthe Colemans were to establish residence in Spain’). Since they are open-ended andcontext-dependent, they are impossible to allocate to a specific category.

The reliability of this coding scheme was measured using the kappa coefficient (K).The kappa coefficient (Carletta, 1996; Cohen, 1960) measures pair-wise agreementamong coders making category judgments, correcting for expected chance agreement.Good quality annotation of discourse or language data normally yields a K of about 0.80(Carletta, 1996). Two linguists (not including the author) acted as coders and categorized2700 of the 3100 samples in the Nissim and Markert (2005) database, containing 984instances that both linguists marked as expressions related to companies. The K wasmeasured at 0.894 (N = 2700; K = 2), which suggests that the coding scheme and theactual coding is reliable.

For the analysis, I focus on the 984 samples which both linguists had marked ascompanies, and three other cases where one of them had originally not understood thecontext (and had therefore not recognized them as companies) but could understand thesamples after the joint discussion between them. This yielded a total of 987 instances ofcompany names. In 20 cases the two linguists could not agree on the reading even afterdiscussion. In the remainder of the paper these are not considered, yielding a corpus of967 coded instances of company names. The distribution of readings of expressionsinvolving these company names is given in Table II.

RESULTS

The primary objectives of the present study were to identify the incidence of metonymyin a corpus of talk involving company names, to document patterns, if any, in the kindof metonymies used, and to reveal whether and how these metonymies connect withmetaphorical expressions.

J. P. Cornelissen88


The Incidence of Metonymies

Through the application of the coding scheme, 295 different metonymies were identifiedwithin 967 instances of company names (see Table II); for a total proportion of over 30per cent. These numbers suggest that the use of metonymies is significant, althoughliteral uses of company names (as in examples (1) and (2)) are prevalent. The use ofcompany names as standing in for something else such as its members or products is, itappears from the data, guided by metonymic compression where the company namefeatures as shorthand for a relative clause or propositional phrase (e.g. ‘Shell’ for‘members of Shell’). In other words, these metonymies appear to have an importantreferential function. Each metonymy features as shorthand for a relative clause orpropositional phrase such as ‘products made by Nokia’. Such clauses or propositions areconverted into metonymies in natural talk.

In short, the answer to the first research question about the incidence of metonymiesfor organizations in natural talk is that indeed metonymies are systematically used bypeople to refer to organizations and reason about them. This systematic use suggests thatmetonymies are not arbitrary single expressions but reflect a general way in which peopletalk about and understand organizations by using one well-understood aspect of acompany to stand for the thing as a whole or for some other aspect of it (Gibbs, 1994;Lakoff and Johnson, 1980, 1999).

Categories of Metonymies

When the 295 metonymies involving company names were coded in specific cate-gories (organization-for-members, organization-for-product, organization-for-facility,organization-for-index, organization-for-event, object-for-name, object-for-represen-tation and other), it became apparent that the most widely used category of metonymies

Table II. Distribution of readings in the corpus (967 instances ofcompany names)

Reading Frequency % of all % of metonymies

Literal 622 64.3 n/aMixed 50 5.2 n/aMetonymies 295 30.5 100

org-for-members 188 19.4 63.7org-for-product 66 6.8 22.4org-for-facility 14 1.4 4.7org-for-index 6 0.6 2.0org-for-event 1 0.1 0.3object-for-name 6 0.6 2.0object-for-representation 1 0.1 0.3Other 13 1.3 4.4

Note: All percentages are rounded to the first decimal.



involving company names is the organization-for-members substitution (Table II). Here,a company name stands in for members of the organization, in such a way that themetonymy leads to a personification of an entire company as carrying out certainbehaviours and as communicating akin to an individual.[5] The following examplesillustrate the organization-for-members metonymy.

(12) BT can be pleased that it bought into McCaw at a much lower price than today’s going rate.

(13) Barclays Bank announced it is preparing to sack 20% of its domestic employees.

(14) A more likely scenario is that GM and Ford will vie for a strategic yet friendly investment in

Jaguar. That would certainly suit the Jaguar management’s book and has less potential for

political embarrassment than an unruly hostile bid.

The result is that in these examples a ‘corporate rhetor’ (Cheney, 1991) is seen toemerge, and indeed the company as a whole is imbued with a certain ‘corporatepersonality’ or ‘corporate identity’ (e.g. Christensen and Cheney, 1994; Czarniawska-Joerges, 1994) as manifested in its collective behaviours and communications. While theinitial motivation for the metonymy may have been primarily referential as shorthand fora relative clause, the use of this kind of metonymy also cues a metaphorical image of acompany as a person or human being. The interesting and significant issue, and one thatI address below, is whether this particular metaphor is antecedent to or rather the effectof the use of an organization-for-members metonymy.

Most of the other categories of metonymies are referential in the sense describedabove, including the organization-for-product and organization-for-facility metonymies.In organization-for-product metonymies, company names reference their products as inexamples (15) and (16) below. This is perhaps only relevant to those kinds of situationswhere the company name is badged onto products and services (so-called monolithic andendorsed branding structures), and not to companies where corporate names are dis-connected (in branding and communications) from its products and services (as in thecase of, for example, Unilever and Procter&Gamble) (Cornelissen, 2004). However, forthose companies to which this metonymic compression applies, it appears that in theminds of managers, consumers and the general public alike it is commonplace toassociate products and services with the company that produces them, and, because ofthe metonymic compression, to substitute these with the company name.

(15) At eleven o’clock the big Volvo appeared at the far end of the hangar and drove slowly towards

him, coming to a stop with its engine running forty feet away.

(16) She gulped down the proffered glass of Coca-Cola quickly.

In organization-for-facility metonymies, a company name refers to the premises orfacilities of the company in question. As such, the usage of this kind of metonymy has asimple referential or dispositional purpose in that it specifies the kind of facilities or thelocation of these facilities for the organization involved. Examples (17) and (18) illustratethis.

(17) Details of crunch talks on a Sainsbury planned for Darlington have been revealed by

Transport Minister Roger Freeman.

J. P. Cornelissen90


(18) The last time he had been in Athens the group that called themselves November 17 had hit

Procter&Gamble with an anti-tank rocket.

The organization-for-index category refers to the metonymic usage of companynames as referencing their stock or shares, and the movements of that stock or shares onan index like the London Stock Exchange. Examples (19) and (20) illustrate this category.The metonymy seems to be centred on a substitution of a company name for stock inorder to easily identify and name the stock involved, but it also leads to a general view ofcompanies as moving up and down an index (and as moving up and down in creditratings). As such, these metonymies are connected to a spatial metaphorical understand-ing of an index like a stock exchange. Spatial metaphors have as their source domain thedistribution of movement or objects (here: company names and they stock they repre-sent) in space (here: an index) (Lakoff and Johnson, 1980).

(19) Fiat, which rose as much as 20 per cent last week on market rumours of outside concerns taking

a stake, fell some three per cent on the day on the Milan stock exchange.

(20) Cheery figures from both Tesco and Next put consumer-related stocks on a firm footing from the

outset, with J Sainsbury adding 6 to 386p, Kingfisher up 2 to 463p and Argos rising 3 to

234p.

Finally, the object-for-name category features a metonymic usage where the companyname is used as a mere signifier rather than referentially, as in examples (21) and (22). Itis a kind of usage where the company name is not used as a substitute for something else(like its products or members) but rather as a signifier or reference point to talk about thecompany, not directly of it.

(21) In the computer industry the power of trademarks can readily be seen. IBM, Wordstar, Lotus

1-2-3 and BBC computer have become household names.

(22) Despite signs of renewed interest from both Ford and Volkswagen (keen to reestablish itself as the

largest car manufacturer in Europe), the British government announced in March 1988 that it

intended to pursue a wholly British solution for the remaining parts of Rover by selling it to

British Aerospace (BAE).

Taken together, the metonymic usage of company names is largely restricted to a fewbasic and conventional ways, with the identified categories of metonymies suggestinglexicalized base interpretations for company names (when used as metonymies). Indeed,the metonymic reference shifts (e.g. organization-for-product) in these categories occursystematically in relation to company names. One explanation for this systematicity isthat metonymies involving company names reflect at least in part lexicalized interpre-tations (permitting reference shifts) for the noun organization. That is, an organization istypically reduced to salient or typical parts such as products, its members or its buildings(Lakoff, 1987; Manning, 1979; Morgan, 1983, 1996), and metonymy is therefore acommon trope in the language of academics, writers, journalists, managers and employ-ees and other members of the general public.

This is a significant finding as it provides support for the thesis that an important partof how people refer to and understand organizations is through specific metonymies. The



use of these metonymies is also conventionalized and entrenched across genres andcontexts of language use; i.e. shared by users of the English language. This does not meanthat people cannot be creative about their use of metonymies or indeed develop metony-mies for company names that move beyond the identified conventional metonymicinterpretations and reference shifts. However, my corpus analysis does suggest that thereare conventional patterns in metonymy usage for company names, and thus conven-tional ways of talking about and understanding these kinds of organizations. They areconventional, having become part of the lexicon, and are used to refer to a general classof organizations, or to the same organization at a number of times and in a number ofcontexts of language use, rather than to one specific instance of an organization. Theseconventional patterns in the use of metonymy led Lakoff (1987; see also Lakoff andJohnson, 1980, 1999) to consider the particular use of metonymies as ‘instances ofgeneral principles’ of language use as ‘they do not just occur one by one’ (Lakoff, 1987,p. 77). Deignan (2005) makes a similar set of observations in her corpus linguistic work,leading her to conclude that ‘hearers do not normally have to reconstruct the metonymicmapping on each occasion, as they may do to interpret non-conventional metonymies.It follows that the semantic link between a metonym and its referent will not be some-thing associated with a specific and perhaps temporary context, but will be a general,relatively permanent and well-known feature’ (Deignan, 2005, p. 58).

Connections between Metonymy and Metaphor

I used Goossens’ (1995a, 1995b) distinction between metonymy-within-metaphor andmetaphor-from-metonymy to interpret the data. Goossens’ distinction is a helpful one forunderstanding metonymy and its central insight, that many expressions are the productof an interaction between metaphor and metonymy, is important for describing andunderstanding language use ( Jakobson, 1990). Interrogating the data in this way, itappears that only two of the metonymy categories (organization-for-members andorganization-for-index) discussed above are connected to metaphorical interpretations oforganizations. All the other categories of metonymy are literal reference-shifts with acompany name standing in for its products, facility, or a company related event, andsignifier-relations with a company name signifying the company as an object or with aname referring to a representation (such as a photo or painting) of the standard referent.An organization-for-members metonymy may cue a metaphorical interpretation of anorganization as a person or human being as in examples (12)–(14) and examples (23)–(25)below.

(23) Motor giant Ford has denied there are more plans for more job cuts at the firm’s Halewood plant

despite lowering car sales.

(24) Intel vociferously denied to Unigram last week suggestions circulated on the internet by GE’s

research and development centre in New York that it has hit a brick wall with its P5 design.

(25) As for NT, Microsoft is back describing it as primarily a server operating system, and

acknowledging that vendors are unlikely to bundle it with more than about 10% of the desktop

machines they ship.

J. P. Cornelissen92


All of these examples hint at an interpretation of an organization as an acting personor human being. This particular metaphorical interpretation is indeed a well-establishedand deep-seated view of organizations (e.g. Baum and Rowley, 2002; Morgan, 2006;Scott, 1998), and has spurned of a whole set of further metaphorical interpretationsof organizations as being able to learn and think as human beings (e.g. Walsh, 1995), toact and communicate (e.g. Cheney, 1991), to have personalities and identities (e.g.Czarniawska-Joerges, 1994) and to evolve, mature and age like them (e.g. Baum, 1996).The fundamental issue here, however, is whether this potential metaphorical interpre-tation of a company as a person is an antecedent or effect of the organization-for-members metonymy. This metaphorical interpretation goes a long way back incorporate history to the logic of granting a corporation with the legal status of a person(e.g. Bakan, 2004; Czarniawska-Joerges, 1994). As early as 1793, one corporate scholaroutlined the logic of corporate personhood when he defined a corporation as ‘a collectionof many individuals united into one body, under a special denomination, having per-petual succession under an artificial form, and vested, by the policy of law, with thecapacity of acting, in several respects, as an individual, particularly of taking and grantingproperty, of contracting obligations, and of suing and being sued, of enjoying privilegesand immunities in common’ (Kyd, 1793, cited in Bakan, 2004, p. 15). By the end of thenineteenth century, courts in the Western hemisphere had fully transformed the corpo-ration into a legal ‘person’. Another strong driver for the widespread use of the meta-phorical interpretation of an organization as a person is the natural inclination in all ofus to ascribe identity traits to an organization, and treat it as (if it were) a person whenwe form an image of it. Cohen and Basu (1987, p. 463) point out in this respect thatstakeholders of an organization are inclined to perceive a company in corporeal terms,and to ascribe traits holistically (where they perceive relationships among features, andalso configural properties beyond merely correlated features, to make up for a perceivedintact entity). They also effectively credit that organization with identity traits, just as theywould an individual person (Cohen and Basu, 1987; see also Gioia et al., 2000).

The metaphor of organization as a person is indeed deep-seated and well established.As such, it may indeed have been formative in our thinking of organizations, and equallyin much of our talking about them. Our linguistic behaviour in the case of organization-for-members metonymies may thus have been informed by it, but the evidence for suchan assertion is far from convincing. For one, most of the identified categories of organi-zational metonymies are not constituted by metaphorical thinking – a clear indicationthat at least not all organizational metonymies are underpinned by metaphors. Moreover,the motivation for organization-for-members metonymies in many instances appears tobe primarily and in the first case referential (as in examples (12)–(14)), as a shorthand fora relative clause or propositional phrase in natural talk, with the metaphorical effectemerging consequently in being cued in the use of the metonymy. This is clearlyillustrated when we paraphrase metonymic sentences such as (12) to explicate thereference shift of the metonymy into ‘senior managers and employees of BT can bepleased that they bought into McCaw at a much lower price than today’s going rate’.Such a paraphrase carries no change in meaning or reference, but importantly does not

cue the metaphorical interpretation of a company as a person. The important point hereis that the metaphor of an organization as a person, one of the most dominant images



within organization theory (Baum and Rowley, 2002), is judged by these sentences ratheras a consequence of a particular way of talking (and thinking) through metonymy aboutorganizations. In other words, the metonymic expressions in the organization-for-members category follow Goossens’ (1995a, 1995b) metaphor-from-metonymy pattern.

The organization-for-index metonymies may cue a spatial metaphor of companies asmoving up and down an index (and as moving up and down in credit ratings) asillustrated with example (11) and examples (26)–(28) below.

(26) LVMH fell 18 per cent to £101 million, and interest costs rose by 26 per cent to £240

million.

(27) IBM was the big loser, dipping 14.5% by value and 3.6% in number of units shipped.

(28) Last time, Shell Transport and Trading came first, Glaxo was second, and Marks and Spencer

came third. This year Marks and Spencer has leapt to the top slot, Shell is now second and Glaxo

third.

The spatial metaphor is indeed present in the understanding of these examples and ofthis category of metonymy. But, again, the evidence for the claim that the metaphor isactually constitutive of, and therefore prior to, the metonymy, is not supported by theresults. If anything, in this category of metonymies the metonymic shift appears prior tothe cueing of the metaphor – the company name substitutes the stock or share first, whichthen in turn is represented through a spatial metaphorical understanding of an indexlike a stock exchange. Similar to the organization-for-members category, then, theseorganization-for-index metonymies follow a metaphor-from-metonymy pattern. Thisfinding is similar to corpus linguistic studies which for certain words or word classes havefound only a few examples of metonymy within metaphor, but dozens of examples ofmetaphor from metonymy (Deignan, 2005; Deignan and Potter, 2004; Goossens, 1995a,1995b). Indeed the lack of instances of metonymy within metaphor may be a reflectionof language use more widely, suggesting that contrary to popular belief in organizationstudies (e.g. Morgan, 1983, 1996; Oswick and Grant, 1996; Oswick et al., 2003),metonymies often give rise to metaphorical interpretations rather than the converse.

DISCUSSION AND CONCLUSIONS

A number of research questions guided this study: (1) What is the incidence of metony-mies in talk about organizations? (2) What different kinds or categories of metonymiesfeature in people’s talk about organizations? (3) To what extent and how are metonymiesconnected to metaphorical expressions concerning organizations? In response to the firstand second questions, I found that a significant percentage (30 per cent) of expressionsaround organizations is metonymical to such an extent that people use a conventional-ized set of metonymies to talk about organizations. In response to the third researchquestion I found that only a few metonymic expressions are connected to (i.e. give rise toor are embedded in) metaphorical interpretations of organizations, and where they were,the linguistic pattern appeared to be one where a metaphorical interpretation wasderived from the metonymic expression rather than the metaphor prefiguring themetonymy.

J. P. Cornelissen94


Hence, the study makes two important substantive contributions to the study ofmetonymy and metaphors within organization studies. First, the study demonstrates thatthere are conventional patterns in the use of metonymies in talk about organizationsincluding frequently used categories of metonymies such as organization-for-productand organization-for-members. In other words, the metonymic reference shifts (e.g.organization-for-product) in these categories occur systematically for company names.This systematicity suggests that metonymies are not arbitrary single expressions butreflect general principles of metonymy use, where people conventionally use salient andwell-understood aspects of a company to stand for the thing as a whole or for some otheraspect of it (Deignan, 2005; Gibbs, 1994; Lakoff, 1987; Lakoff and Johnson, 1980, 1999).Second, the study establishes that many metonymies are not connected to metaphoricalexpressions and that where they are they tend to follow a metaphor-from-metonymypattern rather than a metonymy-within-metaphor one. Only two of the identified cat-egories of metonymies for organizations, namely the organization-for-members and theorganization-for-index categories, may cue further metaphorical interpretations of anorganization. In the first category an organization is personified; in the second it isobjectified as an entity moving spatially. For both cases, there is little evidence to suggestthat these metonymies are constituted by these metaphorical interpretations (i.e.metonymy-within-metaphor). In fact, the metonymic shift appears prior to the cueing ofthe metaphor (i.e. metaphor-from-metonymy). Thus, the data provide little support forthe claim made by Morgan (1980, 1983, 1996, 2006) that metaphor is the ‘master’ tropewithin language (and theorizing) about organizations. This is not to say of course thatmetaphors do not feature within talk about organizations, or that they are not significant.However, judging by the data, it may be that metaphors do not have the omnipresentimpact as has been suggested, or that they simply do not prefigure metonymies aspreviously argued. Another interpretation may be that metaphors are primarily cued inlinguistic and grammatical structures outside of metonymies such as extended phraseswhere a company name or a noun like organization is combined with verbs and/orprepositions. There is a significant tendency within the English language to place meta-phoricity in the verbs (e.g. ‘an organization adapts . . .’) (Cameron, 1999; Sandelandsand Drazin, 1989) and in combinations involving prepositions (in, between, over, etc).Phrases with prefixes tend to cue metaphorical interpretations of organization; forexample, the preposition ‘in’ as in ‘decision-making in organizations’ cues the ratherdominant interpretation of organization as a ‘container’ (see Putnam and Boys, 2006;Putnam et al., 1996).

Implications for Research

This study has a number of important implications for future research within organiza-tion studies on metonymy, metaphor and language more generally. The most obviousone involves the role of metonymy as a common, if not ‘the most common’ (Manning,1979, p. 662), trope in people’s language about organizations. In the study, I foundstrong evidence for the role of metonymies in people’s talk about organizations. In otherwords, metonymies are consistently used by people to ‘talk organizations into existence’(Manning, 1979; Sandelands and Srivatsan, 1993; Taylor and Van Every, 2000; Weick,



2006); to give meaning to organizations and to understand them as particular kinds ofentities. This pattern suggests a clear set of questions for research on metonymies. Whatrole do metonymies play within language use across different contexts, including indi-viduals who work within organizations (CEO, senior managers, middle managers,employees) and academic researchers who study them? What is the significance ofmetonymies vis-à-vis other speech acts and the other tropes of metaphor and ironywithin language use and sensemaking? The present study suggests a ‘strong’ use ofmetonymies in people’s talk with metonymies being used independently of the meta-phors or in many cases actually triggering them, rather than the other way around.Although the corpus of talk extracted from the BNC precludes us from drawing strongerconclusions about the link between the metonymic and metaphoric in our languageabout organizations, it nevertheless highlights interesting questions for further research.

Another implication involves researching and analysing metonymies and metaphorsin our (spoken and written) language about organizations. Previous research onmetonymy and metaphor has been restricted to local levels of language use where thefocus is on the use and meaning of metonymies or metaphors in particular contexts (e.g.Hamilton, 1997; Manning, 1979; Musson and Tietze, 2004; Putnam, 2004; Watson,1995). In contrast, the present research has shown the value of a more global or general

approach to language use across different contexts where the focus is on identifyinggeneral uses of metonymies and metaphors. Within the linguistics community, the last fewyears have seen a rise in the use of electronic corpora for such global research on tropessuch as metaphor and metonymy (Deignan, 2005). In line with this trend, I have tried toshow the value of such a corpus linguistic approach within research on organizationalmetonymies which has allowed me to identify patterns in the use of metonymies whichI would not have been able to with a local analysis of talk in a particular context. I wasalso able to compare the frequencies of different metonymies in the corpus data and tosystematically analyse the connection between the metonymic and metaphoric in peo-ple’s talk about organizations. This particular strength of a corpus linguistic approach isalso its main weakness; in an attempt to identify general uses of language such asmetonymic expressions across contexts of speakers it glosses over pragmatic and contex-tual dimensions of a particular expression within such contexts. Nonetheless, futureresearch may benefit from a corpus linguistic approach to the study of language andtropes. It could be used as an approach on its own when a study is aimed at identifyinggeneral uses of tropes or certain expressions across contexts of speakers, or alongside amore contextual and ‘local’ analysis of language uses. The latter approach would connectcontextual interpretations of the use of a particular trope or expression with an analysisof the conventionality and general usage of that trope or expression across the Englishlanguage which would aid a researcher in making interpretations about its local mean-ings and usage.

ACKNOWLEDGMENTS

I am grateful to Katja Markert and Malvina Nissim for making their data available (see http://homepages.inf.ed.ac.uk/mnissim/mascara/) and for their comments. I also thank Timothy Clark and threeanonymous reviewers for their extremely insightful feedback. The present paper is supported with a grantfrom the Economic and Social Research Council (ESRC) (RES-000-22-0791) to the author.

J. P. Cornelissen96


http://homepages.inf.ed.ac.uk/mnissim/mascara


NOTES

[1] One boundary condition of the study, therefore, is that it is restricted to the English language. Althoughlinguists have examined the conventionality of classes of metonymy and metaphor across differentlanguages (e.g. Lakoff and Johnson, 1999), the present study is focused on, and therefore limited to,metonymies in talk about organizations within spoken and written English.

[2] As a boundary condition I focused on Fortune 500 firms, which is of course not exhaustive of all types oforganizations (e.g. SMEs, public organizations, etc) and of the ways in which people refer to them in talk.

[3] The Nissim and Markert database and their coding scheme can be accessed at http://homepages.inf.ed.ac.uk/mnissim/mascara/.

[4] The British helicopter company Westland was involved in a 1980s economic and political scandal whenLeon Brittan authorized the leaking of a critical letter of Michael Heseltine, the Defence Secretary.

[5] There is an alternative reading for these kinds of expressions which suggests that rather than involvinga substitution of the whole for the part (as in expression (12), for example, where the whole company isevoked as standing in for a specific spokesperson or individual members of the organization), companiesor organizations actually do have an agency in and of themselves that is invested in their members,documents, machines, artifacts as well as in their legal status as persons (Cooren, 2004; Cooren et al.,2005). However, this is in my view not an alternative reading as such an agency is motivated and realizedby metonymic mappings where the agency of an organization is embodied or instantiated in its partssuch as its legal status. These metonymic mappings have now become conventionalized and are often nolonger recognized as such.

REFERENCES

Alvesson, M. (1993). ‘The play of metaphors’. In Hassard, J. and Parker, M. (Eds), Postmodernism andOrganizations. London: Sage, 114–31.

Alvesson, M. and Kärreman, D. (2000). ‘Varieties of discourse: on the study of organizations throughdiscourse analysis’. Human Relations, 53, 1125–49.

Astley, W. G. and Zammuto, R. F. (1992). ‘Organization science, managerial practice, and language games’.Organization Science, 3, 443–60.

Bakan, J. (2004). The Corporation: The Pathological Pursuit of Profit and Power. London: Constable.Baum, J. A. C. (1996). ‘Organizational ecology’. In Clegg, S. R., Hardy, C. and Nord, W. R. (Eds), Handbook

of Organization Studies. London: Sage, 77–114.Baum, J. A. C. and Rowley, T. J. (2002). ‘Organizations: an introduction’. In Baum, J. A. C. (Ed.), Companion

to Organizations. Oxford: Blackwell, 1–34.Brown, R. H. (1977). A Poetic for Sociology: Toward a Logic of Discovery for the Human Sciences. Chicago, IL:

University of Chicago Press.Burke, K. (1969). A Grammar of Motives. Berkeley, CA: University of California (original work published 1945).Cameron, L. (1999). ‘Identifying and describing metaphor in spoken discourse data’. In Cameron, L. and

Low, G. (Eds), Researching and Applying Metaphor. Cambridge: Cambridge University Press, 105–32.Carletta, J. (1996). ‘Assessing agreement on classification tasks: the kappa statistic’. Computational Linguistics,

22, 249–54.Cheney, G. (1991). Rhetoric in an Organizational Society: Managing Multiple Identities. Columbia, SC: University of

South Carolina Press.Christensen, L. T. and Cheney, G. (1994). ‘Articulating identity in an organizational age’. Communication

Yearbook, 17, 222–35.Cohen, J. (1960). ‘A coefficient of agreement for nominal scales’. Educational and Psychological Measurement, 20,

37–46.Cohen, J. B. and Basu, K. (1987). ‘Alternative models of categorization: toward a contingent processing

framework’. Journal of Consumer Research, 13, 455–72.Cooren, F. (2004). ‘Textual agency: how texts do things in organizational settings’. Organization, 11, 373–

93.Cooren, F., Fox, S., Robichaud, D. and Talih, N. (2005). ‘Arguments for a plurified view of the social world:

spacing and timing as hybrid achievements’. Time & Society, 14, 263–80.Cooren, F., Thompson, F., Canestraro, D. and Bodor, T. (2006). ‘From agency to structure: analysis of an

episode in a facilitation process’. Human Relations, 59, 533–65.





Cornelissen, J. (2004). Corporate Communications: Theory and Practice. London: Sage.Cornelissen, J. P. (2005). ‘Beyond compare: metaphor in organization theory’. Academy of Management Review,

30, 751–64.Cornelissen, J. P. (2006). ‘Metaphor and the dynamics of knowledge in organizational theory: a case study

of the organizational identity metaphor’. Journal of Management Studies, 43, 683–709.Czarniawska-Joerges, B. (1994). ‘Narratives of individual and organizational identities’. Communication Year-

book, 17, 193–221.Deignan, A. (2005). Metaphor and Corpus Linguistics. Amsterdam: John Benjamins.Deignan, A. and Potter, L. (2004). ‘A corpus study of metaphors and metonyms in English and Italian’.

Journal of Pragmatics, 36, 1231–52.Eisenhardt, K. M. and Bhatia, M. M. (2002). ‘Organizational complexity and computation’. In Baum,

J. A. C. (Ed.), Companion to Organizations. Oxford: Blackwell, 442–66.Fass, D. (1997). Processing Metaphor and Metonymy. Stanford, CA: Ablex.Foss, N. (2003). ‘Selective intervention and internal hybrids: interpreting and learning from the rise and

decline of the Oticon spaghetti organization’. Organization Science, 14, 331–49.Gibbs, R. W. Jr (1994). The Poetics of Mind: Figurative Thought, Language, and Understanding. Cambridge:

Cambridge University Press.Gioia, D. A., Schultz, M. and Corley, K. G. (2000). ‘Organizational identity, image and adaptive instability’.

Academy of Management Review, 25, 63–81.Goossens, L. (1990). ‘Metaphtonomy: the interaction of metaphor and metonymy in expressions for linguis-

tic action’. Cognitive Linguistics, 1, 323–40.Goossens, L. (1995a). ‘Metaphtonomy: the interaction of metaphor and metonymy in expressions for

linguistic action’. In Goossens, L. et al. (Eds), By Word of Mouth: Metaphor, Metonymy and Linguistic Actionin a Cognitive Perspective. Amsterdam: John Benjamins, 159–74.

Goossens, L. (1995b). ‘From three respectable horses’ mouths: metonymy and conventionalization in adiachronically differentiated database’. In Goossens, L. et al. (Eds), By Word of Mouth: Metaphor, Metonymyand Linguistic Action in a Cognitive Perspective. Amsterdam: John Benjamins, 175–204.

Grant, D. and Oswick, C. (1996). Metaphor and Organizations. London: Sage.Hamilton, P. M. (1997). ‘Rhetorical discourse of local pay’. Organization, 4, 229–54.Jakobson, R. (1990). ‘Two aspects of language and two types of aphasic disturbances’. In Waugh, L. R. and

Monville-Burston, M. (Eds), On Language/Roman Jakobson. Cambridge, MA: Harvard University Press,115–33 (original work published 1956).

Lakoff, G. (1987). Women, Fire, and Dangerous Things. Chicago, IL: University of Chicago Press.Lakoff, G. and Johnson, M. (1980). Metaphors we Live by. Chicago, IL: University of Chicago Press.Lakoff, G. and Johnson, M. (1999). Philosophy in the Flesh: The Embodied Mind and its Challenge to Western Thought.

New York: Basic Books.Levin, F. M. (1991). Mapping the Mind. Hillsdale, NJ: The Analytic Press.Manning, P. K. (1979). ‘Metaphors of the field: varieties of organizational discourse’. Administrative Science

Quarterly, 24, 660–71.Mauws, M. K. and Phillips, N. (1995). ‘Understanding language games’. Organization Science, 6, 322–34.Morgan, G. (1980). ‘Paradigms, metaphors and puzzle solving in organizational theory’. Administrative Science

Quarterly, 25, 605–22.Morgan, G. (1983). ‘More on metaphor: why we cannot control tropes in administrative science’. Adminis-

trative Science Quarterly, 28, 601–7.Morgan, G. (1996). ‘Is there anything more to be said about metaphor?’. In Grant, D. and Oswick, C. (Eds),

Metaphor and Organizations. London: Sage, 227–40.Morgan, G. (2006). Images of Organization, updated 2nd edition. Beverly Hills, CA: Sage.Musson, G. and Tietze, S. (2004). ‘Places and spaces: the role of metonymy in organizational talk’. Journal

of Management Studies, 41, 1533–8.Nissim, M. and Markert, K. (2005). ‘Learning to buy a Renault and talk to BMW: a supervised approach to

conventional metonymy’. International Workshop on Computational Semantics (IWCS2005), Tilburg, 2005.Nunberg, G. (1995). ‘Transfers of meaning’. Journal of Semantics, 12, 109–32.Oswick, C. and Grant, D. (1996). ‘The organization of metaphors and the metaphors of organization: where

are we and where do we go from here?’. In Grant, D. and Oswick, C. (Eds), Metaphor and Organizations.London: Sage, 213–26.

Oswick, C., Keenoy, T. and Grant, D. (2002). ‘Metaphor and analogical reasoning in organization theory:beyond orthodoxy’. Academy of Management Review, 27, 294–303.

J. P. Cornelissen98


Oswick, C., Keenoy, T. and Jones, P. J. (2003). ‘Rethinking organizational metaphors: beyond (M)organi-zation theory’. In Muller, A. and Kieser, A. (Eds), Communication in Organizations: Structures and Practices.Frankfurt: Peter Lang.

Oswick, C., Putnam, L. L. and Keenoy, T. (2004). ‘Tropes, discourse and organizing’. In Grant, D., Hardy,C., Oswick, C. and Putnam, L. L. (Eds), Handbook of Organizational Discourse. London: Sage, 105–27.

Phillips, N. and Hardy, C. (2002). Discourse Analysis: Investigating Processes of Social Construction. London: Sage.Poole, M. S. and Van de Ven, A. (1989). ‘Using paradox to build management and organization theories’.

Academy of Management Review, 14, 562–78.Putnam, L. L. (1995). ‘Formal negotiations: the productive side of organizational conflict’. In Mayden

Nicotera, A. (Ed.), Conflict and Organizations: Communication Processes. New York: State University of NewYork Press, 183–200.

Putnam, L. L. (2004). ‘Dialectical tensions and rhetorical tropes in negotiations’. Organization Studies, 25,35–54.

Putnam, L. and Boys, S. (2006). ‘Revisiting metaphors of organizational communication’. In Clegg, S. R.,Hardy, C., Lawrence, T. B. and Nord, W. R. (Eds), The Sage Handbook of Organization Studies. London:Sage, 541–76.

Putnam, L. L. and Fairhurst, G. T. (2001). ‘Language and discourse in organizations’. In Jablin, F. M. andPutnam, L. L. (Eds), The New Handbook of Organizational Communication. Newbury Park, CA: Sage, 78–136.

Putnam, L., Phillips, N. and Chapman, P. (1996). ‘Metaphors of communication and organization’. InClegg, S., Hardy, C. and Nord, W. (Eds), Handbook of Organization Studies. London: Sage, 375–408.

Sandelands, L. E. and Drazin, R. (1989). ‘On the language of organization theory’. Organization Studies, 10,457–78.

Sandelands, L. E. and Srivatsan, V. (1993). ‘The problem of experience in the study of organizations’.Organization Studies, 14, 1–25.

Scott, W. R. (1998). Organizations: Rational, Natural and Open Systems, 4th edition. Englewood Cliffs, NJ: PrenticeHall.

Searle, J. R. (1979). ‘Metaphor’. In Ortony, A. (Ed.), Metaphor and Thought. Cambridge: Cambridge Univer-sity Press, 92–123.

Sinclair, J. (1991). Corpus, Concordance, Collocation. Oxford: Oxford University Press.Taylor, J. R. and Cooren, F. (1997). ‘What makes communication “organizational”?’. Journal of Pragmatics,

27, 409–38.Taylor, J. R. and Van Every, E. J. (2000). The Emergent Organization: Communication as its Site and Surface.

Mahwah, NJ: Lawrence Erlbaum Associates.Tsoukas, H. (1991). ‘The missing link: a transformational view of metaphors in organizational science’.

Academy of Management Review, 16, 566–85.Walsh, J. P. (1995). ‘Managerial and organizational cognition: notes from a trip down memory lane’.

Organization Science, 6, 280–321.Watson, T. J. (1995). ‘Rhetoric, discourse and argument in organizational sense-making: a reflexive tale’.

Organization Studies, 16, 805–21.Weick, K. E. (1989). ‘Theory construction as disciplined imagination’. Academy of Management Review, 14,

516–31.Weick, K. E. (2006). ‘Faith, evidence, and action: better guesses in an unknowable world’. Organization Studies,

27, 1723–36.Weick, K. E. and Daft, R. L. (1983). ‘The effectiveness of interpretation systems’. In Cameron, K. S. and

Whetten, D. A. (Eds), Organizational Effectiveness: A Comparison of Multiple Models. New York: AcademicPress, 71–93.



Metonymy in Language about Organizations: A Corpus-Based Study of Company Names

Documents

Transcript of Metonymy in Language about Organizations: A Corpus-Based Study of Company Names