This action might not be possible to undo. Are you sure you want to continue?
Blended spaces and performativity
Abstract Mental Spaces theory, with its sophisticated mechanisms for representing the contents of speech, thought, and perception, opens routes for the exploration of relations between these domains and the ``real'' world. One of the most intriguing of these relations is performativity, de®ned by Searle (1989) as the ability of some descriptions to bring about the described situations in reality. My broader de®nition of performativity treats it as mental-space blending (in Fauconnier and Turner's terminology) wherein structure is transferred from a representing space to the space represented. Performative uses of metaphorical representations (pervasive in ritual and magic) are treated alongside literal cases, using our well-developed system for representing metaphor in terms of metal space blends. This approach oers a general treatment of the ways in which linguistic and non-linguistic representations (e.g., ritual enactments or pictures) are understood to aect reality. Keywords: performativity; mental spaces; blended space; metaphor; ritual; magic.
1. Introduction: Linguistic and nonlinguistic performativity When does representation bring about its own truth? Under the right circumstances, not only With this ring I thee wed but a simple I do can bring a marriage into existence.1 Liability for perjury is caused by saying I swear to tell the truth, the whole truth, and nothing but the truth in the right courtroom setting, since this statement constitutes an oath. And a jury's We ®nd the accused not guilty not only constitutes an act of ``®nding'' or judgment, but also legally causes the defendant to have not-guilty status. In one sense, there is no ``magic'' about such facts: as far as I know, parallels prevail in a wide range of dierent cultures. Yet they
Cognitive Linguistics 11±3/4 (2000), 305±333 0936±5907/00/0011±0305 # Walter de Gruyter
306 E. Sweetser bear a fascinating resemblance to other nonlinguistic cases which have been treated very dierently. I myself do not believe that my physical health can be causally in¯uenced by manipulating a representation of me (e.g., a wax doll), but there are others who strongly believe that such a representation can have a causal link to its original, such that stabbing the doll might result in the death of its counterpart person. Performativity is a term which has recently made its way from its philosophical and linguistic domains of origin to other ®elds such as anthropology and critical theory. The central sense of the word, and the one in which I shall be using it, refers to the phenomenon whereby an apparent description of a speech act ``counts as'' a performance of the relevant speech act: I promise you that I'll be there sounds like a description of promising,2 but can in fact constitute an act of promising to be thereÐunlike You promise me that you'll be there. In interdisciplinary usage, performativity has a fascinating pattern of related senses (some of them barely recognizable to users of the ``traditional'' meanings), but the concepts it refers to are all aspects of the human ability to use representations to in¯uence the world outside the representational system.3 It is my contention that setting linguistic performatives in this broader context can help us to understand them, and to decide what is and is not special about the performative uses of linguistic descriptions. My basic argument will be that there are depictive and performative uses of many kinds of symbolic representations, linguistic and otherwise. The contrast between descriptive or depictive and performative is best understood as a contrast between two kinds of causal relationship between the representation and the represented space. In the depictive use of I promise you that I'll be there (presumably a depiction of a habitual activity of promising), it seems evident that the description is judged by its conformity to the represented space: for example, one can ask whether it is a true or accurate description. In the performative use, however, as many scholars have noted,4 no such questions arise: instead, the described ``real'' world is shaped by the speech act, and the speaker has now made a promise. Let us compare these with examples from another domain, which will be analyzed more thoroughly later in the article. A Christian Western bride who wears a white wedding gown is considered by many to be making a (metaphorical) statement about her virginity or sexual purity; this statement is depictive, in that many wedding guests are apparently upset or oended if a divorced woman wears white at her second wedding. They consider that she is making a false or inaccurate representation of herself by wearing white. But a penitent, in Christian and Jewish traditions, puts on white to (metaphorically) represent spiritual purity, not as a description of his or her actual spiritual state, but as a causal aid to bringing about
you're wrong. But the most important thing is that it is possible: under the right . 1983. but I order you all to leave constitutes an order. I shall also be making use of metaphor theory. The central nugget of Austin's perception was that there is something dierent or special about utterances which perform a speech act by saying explicitly I order you. an unusually sinful person may have more need to put on white and repent than a less sinful person. How can so-and-so put on white for Yom Kippur (or for an adult Christian baptism). No. she didn't. or ask by other meansÐas well as from other statements about non-®rst-person or non-present ordering. and of performativity in particular. I shall make use of the theory of mental spaces (Fauconnier 1985 . but the latter utterance doesn't seem susceptible to this kind of contradiction. 1979. or I request. a general framework capable of describing the full range of performative phenomena.Blended spaces and performativity 307 a state of purity. 1997). I tell you. when she/ he is so sinful? Rather. Furthermore. and will also be relying on recent work on blended spaces (Fauconnier and Turner 1996. 2. Turner and Fauconnier 1995). or asking. or tell. Obviously this only works in certain kinds of cases. as opposed to its depictive use in the case of wedding gowns. an addressee of the former utterance can reply. Searle's subtler de®nition of performativity is more intriguing than Austin's: it centers on the issue of a special possibility. since it is a fascinating fact that metaphorical representations (such as the use of white for purity) share this dual possibility regarding their relationship with the spaces they represent. namely that of doing certain particular kinds of action speci®cally by (or in) describing doing them. which sets them apart from the ones which just order. 1998. A short history of the theory of performative linguistic usage Austin's How to Do Things with Words (1962) brought the word performative to the vocabularies of linguists and philosophers. Searle develops a complex theory of the background of social facts against which it becomes possible to say I christen you Adolphus or I declare you husband and wife. In examining these two kinds of causal relationship between representations and what they represent. telling. 1989). This is more like a performative use of the convention of wearing white to represent purity. and have such ®rst-person descriptive statements constitute eective actions which change the state of the world around the speaker. A viewer could not appropriately say. I don't engage in the speech act of ordering by saying Lucinda ordered Dennis to leave. Austin's perception was developed most prominently by Searle (1969. who gave us a much fuller and more ®nely grained theory of speech acts in general.
In descriptions. with the appropriate social authority. an imperative such as drop that! used to convey an order) can be successful or unsuccessful. and that aspect of reality is our own speech acts. depending on its ®t with the world in question. An umpire.. the word ®ts a real or imagined world. Tolkien's The Two Towers (1954 : 189). R. then. Sweetser circumstances. and so on.g. where the good wizard Gandalf breaks the traitor wizard Saruman's sta (both a symbol and an instrument of his power) by saying Saruman. but it will still have been performed: for example. given this understanding of speech interaction. a court ocial can say This court is now in session and thus bring the court into session. 1) calls the direction of ®t between Word and World. as social conventions may decree. says That's an out and makes it an out by saying that it is. but also things which would come under Searle's ``brute fact'' label. or ordering. On the other hand. an utterance couched in a purely descriptive form. since its nonindicative mood removes the possibility of a purely depictive reading. a speaker can engage in a successful act of christening by describing herself as so doing. are institutional facts (Searle 1969: ch. a very clear example of the performative use of description can be found in J. R. A crucial part of the dierence between descriptive and performative. felicitous . Other aspects of reality vary considerably in their receptivity to being in¯uenced by description. Searle also recognizes that magical or supernatural beings are thought of as having dierent performative abilities from humans: they can eect not only social facts. telling. However. The fact of an out in baseball. The circumstances may not make her action felicitous. all it takes for a speaker to perform one of these basic kinds of speech actions is to describe herself as so doing. but it will still be an order (unlike a christening by a person with no authority to christen. or of a court being in session.308 E. As Searle points out. a directive (e. God's Let there be light in Genesis 1 : 3 is perhaps not Searle's best example. lies in what Searle (1979: ch. and hence can be brought into being by the right socially authorized and authoritative speech act. 2) rather than brute facts. But Searle recognizes that such performative actions are not limited to ®rst-person utterances. A descriptionÐor we might better say. a descriptive statement or assertion Ðmay be true or false. which seem to be basic communicative actions with such general conditions that they are likely to be performed frequently by all speakers. we can always shape one speci®c aspect of reality by describing it. in this respect speech acts like christening (which requires very special social authority and circumstances) are dierent from ones like asking. Normally. an order may be addressed to someone who does not agree that the speaker has the authority to give an order. your sta is broken. which according to Austin and Searle will simply not be a christening). In short.
performative statements are more like directives than they are like descriptive statements. but that the statement depends on the order. This explains the ``®rst-person'' nature attributed to performatives by Austin: only ®rst-person actions Ðand only speech actions. in performing a speech act (I order you to drop that. In many respects. Searle (1989) therefore says that performatives are declaratives. Imperative and subjunctive forms mark linguistic utterances as having ``world to word'' ®tÐas being (unlike depictive statements) intended to bring about a state of aairs rather than describe one. because the world has to ®t the words. because it does not involve a ®t of word to world. and a performative such as I order you to leave is therefore both a statement and an order. Searle and Vanderveken (1985) labeled declaratives the class of statements whose truth is ensured simply by the act of uttering them under appropriate circumstances (e. This meeting is now adjourned ).g.5 For an ordinary speaker. rather than the other way around. But a performative statement necessarily presupposes that the speaker's statement is all that is needed to make the world ®t its contents. In the ®rst chapter of Expression and Meaning (1979: 3±4). and is not guaranteed by the speech act itself. in that they too are attempts to make the world ®t the words. there would be neither truth nor practical value to an attempted performative such as I hereby end all wars. however much it may be guaranteed by other contextual factors. or shaping some other aspect of the situation which I have the special social authority to shape by speaking (I christen you spoken by the appointed ceremonial participant. but it cannot be true or false. of courseÐcan generally be performed by and in describing the performance in question. such statements have both directions of ®t: the words automatically ®t the world. Crucially. But they are a rather dierent kind of attempt. Now Searle further points out that other linguistic entities besides statements can be interpreted variously as to the direction of ®t between word and world. and so on). or I request your explanation). he discusses the contrast between a shopping list and the list made up . That's an out spoken by the umpire. but rather an attempt to make the world ®t the words. A command or a request assumes that the world does not automatically ®t itself to the speaker's words: the addressee's mediating action is needed..Blended spaces and performativity 309 or nonfelicitous. By this he means that the truth value of the statement is ensured by the automatic success of the speech act of ordering which the statement performs in describing it. without special power to aect the world beyond the speaker. since this is unfortunately not something whose truth can be brought about by describing it. how else can a statement be construed performatively? So it is only reasonably used where that assumption makes sense: for example.
and (2) between performing an action by describing it (by performative use . The two physical texts could be identical. by its performance. the represented space is taken as ®tting (being causally in¯uenced or changed by) the representation. and using it performatively. an isolated word like ¯our or butterÐ without any surrounding sentenceÐcan participate in both directions of ®t. then the relation between the spaces is depictive or representational. Sweetser by the detective who follows the shopper around the store and notes all the items bought. depiction. telling the shopper what to buy. Performativity occurs when a form whose unmarked function is depictive is used with the opposite direction of ®t. The act of representation. The distinction between these two cases is not that of performative versus descriptive use. the former can thus only correct an error by returning to the store and changing his purchases. But the example clearly shows that. of what was bought. It is the success or failure of depictive ®t (of the representation to the world) which is described as true or false. it represents an ontologically prior mental space. accurate or inaccurate. while the latter can make a correction by altering the list. This de®nition does not distinguish between linguistic and nonlinguistic representations: my point will be that they can be used in remarkably similar ways. A statement is thought of as having an unmarked depictive function: that is. The shopper's actions are judged by their ®t to his list (or failure to ®t it). but is more like the contrast between a directive (the shopper's list is equivalent to a series of instructions to buy the things listed) and a statement (the detective's list is equivalent to a series of statements that the shopper bought the things listed). on the other hand.310 E. while the detective's list is intended to ®t the shopper's actions. and performativity My very general de®nition of performativity is that it involves a particular relation of ®t between a mental space which is a representation. If.6 However. then the relation is performative. in context. but the detective's list is a description. constitutes (or performs as a causal agent in) the structure of the represented space. and the corresponding represented space. This distinction between directions of ®t translates in my analysis into the distinction between two possible directions of causal and ontological relations between two mental spaces. I inherit from Searle the crucial distinctions (1) between using a representation descriptively. and are thus ontologically and causally prior to it. true or false. 3. while the shopping list is a self-addressed directive. If the representation is taken as ®tting the represented space. where the words bring about the described world state. Mental spaces.
In the painting. and close personal possessions are similarly used in voodoo). Metonymic force may be added to such a literal representation: for example. In using a name to refer to a present entity or add to a depictive mental space involving a non-present entity. Frequently the link between the depiction and the world it is intended to aect is asserted by metonymic means. I may order you to leave by pointing to the door. bualo fall prey to successful hunters. or by using a linguistic form which is appropriate to the speech act of ordering (an imperative such as Leave this room! ). rather than depictive? As cultural descriptions everywhere attest. we all use a wide variety of representations both depictively and performatively. to make sure that the representation resembles me. whether linguistically or not.Blended spaces and performativity 311 of description) and performing it by other means. both of these perform the relevant act directly. you might paint a picture of me in graduation robes. discovered by archaeologists who know from associated artifacts that the social group which produced the paintings were also bualo hunters. In attempting to bring this about by the performative use of purely representational means. Names seem to be like other representations in being susceptible to referential use for either depictive or performative purposes. Linguistic representations are only one strand (though an important and complex one) in the web of mental spaces which show such relationships. one might well use metonymic links such as dressing the human actor playing the role of the hunted animal in the hide and/or horns of an animal of the same species. is it performative. and put a lock of the actual student's hair on its head (hair. which is a performative use of a descriptive form. I can convey the order by saying I order you to leave the room. Was the painter chronicling a successful past huntÐ is it a depictive record. suppose that you want to magically bring it about that I pass my exams and graduate from college. On the other hand. A nonlinguistic example to start our discussion might be that of a painting of a bualo hunt on a cave wall. One might try to bring about a successful hunt by painting such a scene. following and modeling itself on events in the represented world? Or was it made with the intention of bringing about success in a future huntÐis it intended to magically bring about in reality the situation it represents? That is. in the latter case. as well as by purely depictive means. Nonlinguistic performative examples abound in ritual and magic. the kind of reference involved is like depictive reference in that it re¯ects already established naming practices . A possible debate about its proper interpretation might run as follows. blood. nail ®lings. with diploma in hand. For example. one could dress a wax doll in graduation robes. or by enacting one ritually. or paste a photograph of my face into a graduation picture of this kind.
it is not metaphorically describing an extant or possible situation. Metonymic performative use of names is possible as well. lest that bring them into the place where the namer is. and hence GAINING STATUS IS RISING.g.7 Names of gods are powerful because they may invoke the presence of the deity. Russian m'edv'et' ). A normal depictive referential use of a name brings neither the named entity nor the naming convention into being or presence. But unlike most metaphors. 4. One is involved in the use of names as invocations or evocations of the named entity. it used to be the custom to carry a newly born infant up a ¯ight of stairs as soon as possible after birth.. the relations in the target domain (status) are to be changed. we just mentioned the traditional belief that even purely referentially-intended mention of a powerful entity evokes the presence of that entity. By changing relations in the source domain (height). For example. and follows the presumed naming convention to do so. being replaced in Germanic by words meaning ``brown one'' such as bear. George Lako tells me that in some Italian village communities. Names of evil spirits are generally taboo. so that the child might socially ``rise in the world'' in later life. and in Slavic by words which mean ``honey-eater'' (e. lest their use invoke the spirit's presence. Sweetser and makes use of them to depict some entity in some space. but silver baby spoons try to make itÐand the wealth with which it is metonymically associatedÐcome true. Perhaps .312 E. The IndoEuropean name for ``bear'' (the ancestor of Latin ursus) appears to have been lost in some Indo-European subfamilies. Metonymic relations are similarly capable of being used either depictively or performatively. but attempting to in¯uence a future state of aairs. by use of a new name. The metaphorical mapping seems clear: STATUS IS UP (cf. The second kind of performative use of names is the metalinguistic one which establishes a naming convention by using a name: ``baptism''. A metonymic ritual which parallels the ``Baby and the Stairs'' ritual just mentioned would be the rather common European and American practice of feeding a baby with a silver spoon. But two kinds of performative naming practices seem possible. Ritual and performativity A particularly interesting fact about the dual nature of representations discussed above is that it holds true of metaphorical representations as well as of literal ones. metonymically invoking the frame of prosperity. Lako and Johnson 1980). The phrase born with a silver spoon in his mouth seems hard to imagine as a literal description. but represents the being which is presumed to exist. so to speak. This seems to have been due to a taboo on naming certain powerful (probably totemic) animals.
and allow it to be erased by their walking. or the memory of him. and paints eyes on its prow: this can only be metaphorical. or metaphorically depict his independently determined nature. and his ®ercely independent character. often in the same places with tablets which bear written curses.) The possibility then exists of ``®t'' in either direction between these two spaces. Similar magical uses apparently prevailed in late classical Europe. In Ursula Le Guin's A Wizard of Earthsea (1968). and the other (the target) is the space of the represented world. Metaphorical and metonymic representations. so the metaphorical use is a secondary representational use of it. since the boat cannot literally see. of course. the thing being represented is the social or intellectual perceptiveness. but ironically one must remember him to make sure he is forgotten. one of these (the source domain) is the space of the representation. other characters. At Purim. by foiling the plot of Haman. He names his boat Lookfar. The holiday commemorates Esther's saving of the Jewish people. can be used either depictively or performatively.8 . (This description is itself a representation. is metonymically obliterated by erasing or destroying a written copy of his name. the central character is named Sparrowhawk originally by metonymic association (as a young magic-worker he loves to summon animals magically and is especially often seen with hawks). these tablets were dropped into wells as a way of conveying them to the underworld deities who might be able to carry out their harmful intents. since it is intended to give the boat some of the protection that comes from careful ( far-sighted ) physical observance and mental consideration of the environment. For instance. such as his ®rst teacher. He. but it is also performative rather than depictive. King Ahasuerus' evil counsellor. judging by the ®nding of pierced and torn lead foil tablets. the religious holiday when Jews read the Book of Esther. just like literal representations. and the representation of it is the description of physical vision. then. with victims' names written on them. the source and target domains. some observant Jews write the Hebrew characters for the name Amalek on the soles of their shoes in chalk. A metaphorical mapping automatically brings two mental spaces into play. Haman should be doomed to oblivion.Blended spaces and performativity 313 a stronger example of metonymic name use occurs in a Jewish Purim ritual. Both of these are depictive uses. in saying clear-sighted or sharp-eyed to refer to a person's social or intellectual perceptions rather than to their physical vision. Haman was a member of the tribe called Amalekites. clearly think the name is a good metaphorical description of his intellectual speed and power. since the name is supposed to metonymically describe him as he independently is. Further. And one traditional rabbinic instruction for celebrating Purim says that Jews should ``remember to forget (or obliterate) the name of Amalek''Ðthat is.
they metaphorically refer to the desired spiritual end-state of the rituals in which they participate. and greatness. or magnify (i. you are glorious. in the physical uniting of the blessed bread and wine (metaphorically representing Christ's body and blood) with the bodies of the worshippers.. glory. and is also intended to assist in maintaining it. However. not simply depictive. in support of this depictive reading we noted the convention barring the wearing of a white dress at a woman's second marriage ceremony. we also mentioned the white ritual garments worn for puri®cation (by penitents in both Jewish and Christian traditions) as a performative usage of the same metaphoric mapping: being intended to help bring about purity. has been interpreted as both acknowledging and perpetuating the special link of that king and his people to that territory. an already extant spiritual union between human and divine.. We cited earlier the white dress worn by many Christians as metaphorically depicting (truly or not) the bride's virgin status. including Indo-European ones. is especially certain to be simultaneously depictive and performative. is con®rmed by many wearers' superstitions about taking o or losing their wedding rings. ``make glorious''). since it both acknowledges (or arms) a particular ontology as prevailing. Does kneeling to a divinity metaphorically represent the already extant dierential in power and status between worshipper and god (a depictive use).9 Perhaps ontological religious ritual. Indeed. the same representation in ritual may apparently be simultaneously depictive and performative.e. The apparently traditional Celtic ritual marriage of the king to the land. are now analyzed as depictives where worshippers acknowledge God's blessedness. A Christian communion service may likewise be seen as metaphorically depicting. or you are great. but its use in a wedding ceremony is to bring that permanence into social being.10 Jewish and Christian (following Jewish texts) ritual words which claim to bless. but it certainly must also be seen as intending to causally bring about this spiritual union via the consumption of the bread and wine.e. on the grounds that it is a misrepresentation. ``make great'') God by saying You are blessed. or help to bring the worshipper into the right state of humility (a performative use)? Perhaps both. or keeping it in being.314 E. The ring's status as performative. thought to have been manifested metaphorically in a ritual act of sexual union with his spouse. Sweetser Much religious ritual seems to be both metaphorical and performative in this sense. not the factual initial state. of a more . glorify (i. such as the communion service. The circular shape of a ring metaphorically represents the unending permanence of marriage. not just to describe it. But there is evidence in older Mediterranean traditions.
performatively. and sometimes seem in broader contexts to have the same ambiguity as to their performative or depictive status: does kneeling before a monarch acknowledge. 2000). physical objects such as bread and wine. such as the fact of loom placement in a house. What Bickel brings out is the complex unity between what Westerners would regard as ``everyday'' activities. and actions such as eating or kneeling. An added point of interest is that the relevant representation in ritual can be linguistic or nonlinguistic: it takes place via visual symbols like color of garments. DOWN is relative lack of such status). and ``ritual'' activities such as worship at a temple. Often it seems that wearing a ``lucky'' piece of clothing or jewelry is intended to bring about success associated with the item earlier: a baseball player who pitched a no-hitter wearing a particular hat or socks may insist on wearing the ``lucky'' hat or socks in future important games. and that ritual often involves both metaphorical and literal representation. or both depictively and performatively. LIFE/DEATH (UP is ALIVE. or the right way to pass your grandmother on a hillside path. which may be used depictively. Likewise. where gods depended on worshippers for greatness (and were thus literally made glorious or great by worshippers' depictions in hymns such as those of the Vedas. and GODS/HUMANS. DOWN is DEAD). chronicles its pervasive metaphorical mappings of the source domain of UP/DOWN onto target domains such as STATUS (UP is positive status. temples are built on local heights. and the leaves of food sacri®ces to the gods should face uphill. We should note that not only ``ritual''.12 Metonymic personal good luck charms abound in Western cultural contexts. Bickel (1997. Western customs of bowing and kneeling (to monarchs for instance) are not limited to religious ritual. or help to maintain his or her rule as monarch?11 ``Personal ritual'' in everyday life in Western cultures has many of the same characteristics of more culturally shared rituals. The lucky . leaving the uphill side to the respected elder. The crucial points here are that metaphorical as well as literal description can be used either depictively or performatively (as claimed in the introduction). a loom is placed facing uphill to weave garments for living wearers but downhill to weave a shroud. a grandchild would have to pass a grandparent on the downhill side of the path. In Belhara. but also everyday interaction has the characteristics we have just been mentioning. even while worshippers depended on gods for powerful supernatural protection and support enabled by that same greatness. the Avestas. analyzing Belhare culture. as well as via linguistic representations in the spoken text of the liturgy. Table 1 provides a summary. or the Homeric hymns).Blended spaces and performativity 315 performative interpretation.
a rookie baseball player might possess a hat or a bat once owned by a great player. GLASS CEILING. FAR-SIGHTED policy. Dorian Gray's portrait. again with performative intent to create similar success for the rookie. use of lock of hair to make voodoo doll eective.'' description of social fact in order to portray it: ``it's out'' said by a fan. commemorative reenactment of hunt. wearing or using the lucky item would be intended to blend the great player's past successes with the rookie's actual game situations. metonymic item metonymically evokes the past success frame. description of speech act in order to perform it: ``I promise you to come on time.g.g. Another reason why an item may be linked with success is an association with a successful previous owner. For example. Sweetser Table 1. metaphorical metonymic Linguistic literal metaphorical metaphorical descriptions used to portray: prices SOARED. with the causal direction of the blend being clearly performative. ritual hunt to bring about future success in hunting. still-uncertain game situation. etc. Classes of uses of representations Depictive Nonlinguistic literal Performative ordinary portrait. enactment intended to portray literally: e.316 E. description of speech act in order to portray it: ``He promised me to come on time. wearing white at a wedding. . In this case. metonymic descriptions used to portray: HAND~``worker''. wearing white as a penitent. so wearing it in a subsequent game blends the past successful frame with the ongoing.'' description of social fact in order to create it: ``it's out'' said by an umpire. picture of high-heeled shoe on women's rest room door.. metaphorical ritual formulae intended to change the world: ``This is my body'' in the ritual of the Eucharist. metonymic formulae intended to change the world: saying names to invoke the named entity is a frame-metonymic evocation of the situation of addressing a present entity. enactment intended to in¯uence events: e.
A linguist who uses.Blended spaces and performativity 317 Personal ``luck'' tokens which are not so obviously related to professional activity can also be understood in a similar way. Supposing that the donor is not a member of the relevant professional community. Past anthropological theories have indeed given some of these insights. and if the young linguist also came to associate the pocketwatch with his or her own successes. Any sucient theory of these domains should be able to oer insight into the motivated relationships between ritual activities and their intended eects. let's say. tend to feel happy and strong and con®dent with each other. Roman Jakobson's pocket watch as a ``lucky charm'' when presenting conference papers. and help the ``on-stage'' professional to feel con®dent and to experience the professional environment as supportive. or the baseball player. it seems necessary to relate this account of performative ritual structures to some of the anthropological understandings of magic and ritual. Having come so far. a parent or spouse or lover. or loving parents and children. of course. by metonymically evoking the space of the relationship. why would someone wear such a gift to ensure luck in professional performance? The answer. was a famous linguist. The object given by the spouse or lover or parent is intended to accomplish the same task. is indeed evoking a second space to blend with that of the professional performance: the social space of his or her relationship with the donor. We might suppose that well-paired couples. Now suppose that the young linguist carried the much-loved mentor's pocket-watch to bring success in conference presentations. and blending the associated feelings of success and con®dence with the professional performance setting. but why? A blended spaces analysis would argue that the linguist. A spouse's presence in the audience at a professional event might also blend the two spaces. For example. there would be yet a third source of ``power'' to the lucky charm. These ``lucky'' item traditions are not mutually exclusive: perhaps the most powerful kind of professional lucky charm is one which combines them. ``magic'' and ``superstition'' tend in earlier . imagine a linguist whose beloved teacher. is that it is a source of moral support. For example. is presumably doing much the same thing as the baseball rookie wearing the role model's hat. Two of the kinds of blending input discussed above would then be in force. and feel involved in mutually supportive social interaction. although their helpfulness is unfortunately obscured by a language which is often alien to modern scholarship. or parent. and the item has no metonymic association with success in the relevant ®eld except what the current wearer can bring to it. But a linguist or a baseball player may also attribute ``lucky'' qualities to an item which is a gift from some special person.
and on ``similarity'' or ``analogy'' to them. thus making a link between traditional analyses and modern metaphor theory. . including those on which performative ritual usages are based. showing how mappings can be laid out between source and target domains in many cases. The structuralist enterprise in cultural analysis depends crucially on the hypothesis that not only magic but culture (and hence magic) inherently involves systemic ``parallels'' or structural analogies between disparate domains. The work of scholars like Turner (1967) or Levi-Strauss (1966) consists of constant constructions of analogical mappings. Sweetser Western anthropological and folklore literature to refer to the beliefs and activities of people who are less educated or of non-European cultures. Evans-Pritchard's (1937) study of Azande witchcraft and magic describes. while there is also real metaphorical magic as well. a cure for leprosy which used the bark of a tree which looks like the patchy appearing skin of the early stages of leprosy (1937: 450). metaphor. often via verbal and non-verbal representations used in the manner we have discussed. which are used to explain the relevant culture's causal and ontological theories. and rituals for keeping rain from falling by suspending a stone. while ``religion'' refers to the beliefs and activities of ``advanced'' cultures. Sùrenson (1997. Anthropologists have agreed that the contiguity-based (``contagion'') kind of magic is essentially metonymic. An extensive literature on magic and ritual cannot all be covered here.15 Anthropologists have insisted that magic is based on the community's belief in the ecacy of the practices involved (cf. similarity. for example. Evans-Pritchard 1937. The leprosy treatment seems more similarity based in the direct sense. while the anti-rain magic is more analogical or metaphorical.318 E. which are understood as having causal relations to each other. anthropological analysis has proceeded without a very precise de®nition of analogy. and opposition. and the ``similarity'' kind is metaphorical or analogical. In recent work. 2000) oers a very neat analysis of metaphorical magical structures. shape. and so on. In general.14 Mauss (1950) resumes past work by reference to the laws of contiguity. or by sending smoke upwards (1937: 472±474) in order suspend or reverse the direction of the falling water. or similarity.13 Setting this aside. However. A metaphor analyst would be inclined to say that some ``similarity-based'' magic is based on literal similarity of color. many past analysts have noticed that magic is based on ``contiguity'' with the supposed causal eecting agencies or the aected entities. the activities labeled as ``magic'' are clearly intended to have a causal eect on the world. The leprosy treatment is also contiguity based (hence metonymic) if the bark is brought into contact with the body of the person to be cured.
In the ritual setting. to develop a new general theory of performativity which goes beyond speech acts. It oers a way to take the metaphorical and metonymic links which seem to be the basis for much of magic and ritual. Likewise. or that they . are intended to aect the world around them: he does not apparently mean the word to refer precisely to the use of representations in my Searlean manner. The mental space structures involved in performative and depictive usages Let us ®rst look at the blended spaces involved in the earlier hypothetical example of a ritual dance which enacts a successful bualo hunting scenario in order to bring about success in an intended future real hunt. and to incorporate them into a theory of cognitive structure. mental spaces oers a ready formalism for the kinds of cross-domain relationships (based on perceptions of parallelism) which are dealt with by structuralist analysis of culture. Malinowski 1965. If we add to this the Searlean concept of performativityÐin the sense of the use of a representation to aect the represented spaceÐthen we can bring together our theories of metaphor and of mental spaces with Searle's speech act theory. for example. Figure 1 shows the complex blend involved in the performative example. They have highlighted the importance of magical words in many or most traditions. including verbal formulae. Since cognition at large is structured in terms of spaces and mappings. although some things (such as who is known to be a good hunter in real life) may be brought in. A mental spaces analysis of ritual and magic gives a uni®ed analysis of the structures of the beliefs which permit magic to be seen as ecacious. Tambiah (1985) resumes much of these understandings with a general claim that magic is ``performative'' in Austin's sense. there must be knowledge of the speed at which a bualo runs. and how strong and heavy a bualo is. 5.Blended spaces and performativity 319 Mauss 1950. This theory. so not all of the accessible knowledge about these people and weapons is brought into the ritual interaction input space. and many more). in the input to the hunting scenario. By this he means that magical activities. with weapons. We have a blend between (Input 1) the ritual setting and participants and (Input 2) a hunting scene and its participants. has great potential for incorporating past insights into magic and ritual and unifying them with our understanding of everyday cultural cognition and language. it seems to me. This pursuit is not interpreted as hostility within Input 1. Firth 1954. Some of this is going to be left out of the structure of the hunting scenario which constitutes Input 2: ritual hunters will probably not assume that they need to run fast enough to chase a real bualo. we have a real space16 containing people chasing other people.
for them the person enacting the part of the bualo is a bualo. The bualo-enactors may wear bualo skins. the ritual hunters may be the same people who intend to participate in the planned expedition. and the people enacting the roles of hunters are the human participants in the relevant future hunting expedition. creating metonymic links of a dierent kind between the two spaces. The blend is the eective understanding of what the ritual is. which . will need to remove and cure the (perhaps already cured) bualo hides worn by the bualo-enactors. giving identity links between the two input spaces. there is an added complexity in this hunting ritual blend. for the participants: that is. and may wear their actual hunting gear. This blend may be strengthened in various ways: for example.320 E. As Gilles Fauconnier and Mark Turner have pointed out to me. however. even though these are not the hides of the hunter's actual future quarry. Sweetser Figure 1.
while the physical pursuit of humans by other humans in Input 1 is happening in the time frame of the ritual performance. in a single action over a single time period. Fauconnier and Turner 2000) on multiple dimensions makes it possible to conduct this ritual without taking all the time needed for a real hunt. via the physical act of pronouncing one phrase. that of the baby being carried upstairs to ensure her future success in the world. the description and the performance are not only blended. This is of course a simpli®ed analysis: in a full treatment. Let us now compare the hunting example. in the blended space set up by the ritual. one would also. One way of looking at metaphor is to think of it as a special ``one-sided'' blend of mental spaces. Via this compressed mapping onto the Input 1 space. But many rituals are not as temporally simple as that. If I say I promise to be there. In this case. the entire time span of Input 2 is yet to happen. The ritual hunters' success in chasing down the enacter of the ritual bualo role causally brings about success in the future hunting expedition. which shapes the ritual participants' concept of the future hunting expedition. Compression (cf. one can bring about a result in the target domain which is metaphorically being conveyed via this depiction (IMPROVEMENT IN SOCIAL STATUS). before it can come into existence and set up Input 2. The blends involved become more and more complex. I have described my act of promising and thereby performed a promise. In a performative speech act. with a metaphorical ritual blended space example. Input 2 is causally in¯uenced. still in the ritual participants' anticipatory imaginations. the perhaps much shorter time-span of the ritual is mapped onto the longer time-span of an imagined actual hunt. for example.Blended spaces and performativity 321 we have not discussed. and. they will not be cognitively blending this space with that of the human-hunting-human ritual activity. assuming that many bualo might be killed in an important hunting expedition. but the basic internal structure of . and even without knowing how long the actual hunt will last or how many bualo will be killed. they are simultaneously enacted. one needs to believe that in depicting a source-domain situation (MOVING UPWARDS). presumably the actual hunt will not in fact be a blend of this kind: that is. which is mostly literal and metonymic in its means of depiction. when the hunters on the intended expedition actually kill a bualo. the catching and ``killing'' of the fewer human ritual enactors of the bualo role stands for the intended hunting and killing of many more real bualo. In order to believe in this ritual. speci®cally a blend wherein the structure of the source-domain input space is used to restructure or add structure to that of the target-domain input space. Here. treat the input from past experience of hunting expeditions. because in the blend it constitutes success in the future expedition. Further.
is the source domain. But metaphorical usage minimally involves two input spaces in the representation. Pressure from the blend. Input 2 is the child's life: this brings in anything we know about life-directions. which represents some space wherein physical rising occurs and (2) a target-domain depiction of gain in status. then. Via metaphorical one-sided mapping from source depiction to target depiction. In a metaphorical depictive use. Input 1 is the understanding of this particular act of carrying the baby upstairs: this brings in material from everyday stairclimbing. The more relevant space for the representation part of the performative. a blend is created. she's rising used metaphorically to mean that she is gaining status involves (1) a sourcedomain depiction of physical rising. a literal use minimally involves two spaces. with a positive and negative pole to the scale (since both verticality and status seem to have scalar inference structures and +/7 poles). in an ordinary trip upstairs carrying a baby. Metaphorical performative uses do not pose any particular problems. The blend is the child's life direction AS going upstairs. as well as the spaces represented by each of those representations. including being carried and the causal role of the carrier. including pressure from compression. Finally. and the target domain is taken as crucially connected to the represented and aected space. For example. one might stumble or fall. Thus. which represents some space wherein status is gained. that of the representation and that of the represented situation. including parental in¯uence. Figure 2 shows a partial mental spaces analysis of the ritual of carrying the baby upstairs to ensure her improvement in status during her life. in¯uences how the two input spaces actually structure the resources from the cognitive domains which ultimately provide their material. By this analysis. In this ®gure. and we understand the utterance to refer to a gain in status understood in terms of physical rising. But it might well be an extremely bad omen to stumble or fall in this ritual stair-climbing activity: this is because the baby's entire metaphorical ``life journey'' is represented by these few moments of stair-climbing. and likewise in a metaphorical performative use. Sweetser the blended space is primarily determined by the target domain.322 E. again: the baby's whole life corresponds to the much shorter whole trip upstairs. and that would be unimportant as long as neither the carrier nor the baby were physically injured. the generic space is some very abstract scalar structure. There is compression of structure. it is understood that the blend is a new way of construing the representation of the target space. A single tumble on a short ¯ight of stairs takes up a large portion of timeÐand of . not the source space. it is understood that the target space is the one to be aected by the performative.
and go down again to fetch it (an activity presumably not mapped by the ritual). in the blended space of the ritual the movement upwards causes the future improvement in social status.Blended spaces and performativity 323 Figure 2. Similarly. only selected connections have been drawn. the stair-climbing activity must be continuous from beginning to end. In order to believe in the ecacy of either of these rituals. Once the ritual ``clock'' is started. . and that one can causally bring about improvement in social status by enacting something which is metaphorically mapped onto it. one could not decide in the middle of the ritual that one had forgotten the baby's bottle downstairs. one has to believe that one can causally bring about a good hunt by (or in) enacting it. The importance of the tumble relative to the stairclimbing (not relative to the participants' whole lives) is mapped onto the importance of some misfortune relative to the baby's entire life. just as the baby's life will be continuous from start to ®nish. because it constitutes it.) the watchers' attentionÐrelative to the time and attention spent on the stair-climb as a whole. As in the hunting ritual. (To keep the ®gure legible.
the Catholics dier from the Protestants with respect to the causal nature of the consecration part of the ritual: they see This is my body and This is my blood as being performative. the ``Baby and the Stairs'' ritual is only an attempt at the performative use of a description. rather than by other means. since the whole point of the depictive metaphorical portrayal is to bring about the changes in the world which are metaphorically depictedÐand to bring it about by the depiction. However. These are not minor issues. First of all. We must still see performativity here. Positions run the gamut from those who simply see heterosexuality as the essential cornerstone of the concept of marriage (in which case. To the second set. Austin may be correct that nobody will ever think a marriage formula pronounced over a human and a monkey will ``count'' or ``take eect''. then? It seems clear that there are. same-sex marriages are simply causally inecacious and nonsensical). if successfully carried out. while another might see the ritual as simply an attempt to make such an improvement in status occur. ``Thinking makes it so'' By this analysis. which has little to do . while most Protestant denominations hold instead that the Eucharist is a ``memorial'' service. One set of parents might see the ``Baby and the Stairs'' ritual as having. to those who see heterosexuality as a socially imposed standard on marriage. there are cases whereÐcontrary to Austin's original simpler claimsÐthere may be real disagreement or uncertainty about the social authority which gives causal shaping ability to a particular speaker. One of the most basic dierences between Catholic and Protestant theology is that Catholicism holds that the Eucharistic bread (host) and wine are really transformed into Christ's body in each performance of the Mass. even if it is not viewed as necessarily successful. and the bread represents Christ's body.324 E. already causally set the ``upwards'' socioeconomic trajectory of their child's life. But the negotiability of the authority of the formulaic speech acts involved in marriage is clear from the current lively social debate as to whether same-sex marriage is possible. while the Protestants see these phrases as metaphorically depictive. Are there linguistic ``attempted'' performatives. One might simplistically but insightfully summarize this as a disagreement about the status of the blend created by the eucharistic rite. but is not the Real Presence. performative causal force is generally dependent on the beliefs of the people making and interpreting the representations which are taken as being used performatively. Sweetser 6. Both groups see the Eucharist as performative in that it brings about union with God via a metaphorical blend with the real space consumption of bread and wine.
for example). they succeed by nondescriptive means. might constitute a command.Blended spaces and performativity 325 with the essential idea of marriage as a relationship (in which case samesex marriages are perfectly possible. which will not be recognized by Orthodox Jewish rabbinical authorities. Where such acts dier from nonperformatives is that their causal forceÐsuccessful or notÐis still carried out via apparently descriptive forms. or Reform Jewish marriages between a Jew and a Reform convert to Judaism. The possibility of ``attempted'' performative uses of description puts these performatives on a par with other kinds of nonperformative speech actsÐunlike I promise you. With this ring I thee wed may not always succeed in its causal goal of constituting a genuine marriage vow. civil marriages of Catholics who have only received a civil divorce and not a religious annulment of the ®rst marriage. If that belief is shared by hearers. which always constitute a genuine promise or question or order. or I order you.17 An even simpler case is that of marriages recognized by civil authorities but not by religious ones: for example. or Jones will remain here. then a description may indeed have causal force in this way. it constitutes a command via the power of description. Their grammatical forms do the work that performative description of the speaker's action could also do. and Smith will proceed to the rendezvous point. If so. which are centrally present tense (performed in and via the description of one's own speech act. And this in turn . The presumption that one can describe someone else's future actions may. This follows readily from the observation that through speech one can causally in¯uence one's own present actionsÐespecially speech actionsÐbut only the future actions of others. You will proceed directly to the rendezvous point. which are future and are not centrally about speech acts. but do succeed in their illocutionary goals of directing and querying just as inevitably as their performative counterparts. and verbal formulae of marriage are causally ecacious for same-sex couples). This is unlike Please ®nish your spinach! or Where is Chris? Direct speech acts of this kind may not be causally successful in their perlocutionary goals of in¯uencing the hearer's actions. or I ask you. and second. said by a superior ocer to soldiers Jones and Smith.and third-person ones. in the right context. Another class of linguistic forms which might be treated as part of this class of attempted performativesÐperformatives whose success is not assured by the performance itselfÐis non-®rst-person descriptions of future events which are intended as directives. rather than via the direct use of imperative forms. however.18 There is a fascinating complementarity between ®rst-person linguistic performatives. in itself indicate the belief that one has sucient social authority to cause those actions.
which may not be mine. . I may say ``tomahto'' and you may try to impose your pronunciation on me. And further. which can be quite coercive. that ®ts my independently extant gender identity. constantly. only about their mapping onto labels. Indeed. For example. as well as woman or man. is that the one sets up the alternate gender identity only in the world of the play. A critical theorist might reply to all of this that in fact my gender identity is anything but ``independently extant''Ðit is precisely in doing all the things which I do (such as wearing gender-appropriate clothing) that I continuously enact my gender identity. in this understanding. while cross dressing may attempt to establish it in the real world. has this characteristic. while if I wore other clothes. again without much real disagreement about the entities. But nonetheless. Sweetser provides the solution to the paradox proposed by cultural theory. by this analysis. Part of the reason for this seems to be their questioning of the boundaries between depiction and performativity in social construction. this is not done in isolation. that would be an attempt to create some alternative gender identity.326 E. all of the uses of gender markers are ``performative''Ðthey all construct gender identity. 1993). which treat all descriptive speeches as performatives. and not coincidentally they are also far more contested. The dierence between stage costume and ``cross dressing''. The concepts of femininity and femaleness are far more socially constructed than felinity or stoneness. in particular by analyses such as Butler's (1990. but as part of a complex and inescapable web of social conventions. Of course. nor will we have a changed tomatohood depending on who wins. each use of the word stone or cat. and contributing to the viability of its successors. there are evident reasons why cultural theory has focused on words like woman rather than on ones like stone or cat. but in our discussion there is no disagreement about what a tomato is. Critical theory uses of the word performative seem to have generalized the term to refer to almost any kind of action which in¯uences social construal of a situation or a person. it is my authority as a native speaker of a relatively highprestige dialect of English which may make my labeling practices potentially citable. every linguistic categorization I make is an act justi®ed by its predecessors. Or we may debate about the categories fruit and vegetable. which are eectual only through the ``citation'' of past ones. we might say that many of us have a folk theory that when I wear certain clothes. However. This kind of con¯ict is less likely in less socially constructed areas. as it was in part the social authority of past users which makes me cite their uses. Labeling me a woman may bring with it the knowledge that the labeler is imposing on me his or her understanding of femininity.
I will be most unlikely to aect the situation described: whether a cat is present or not. In the case of one's own social identity. therefore. or even of our understanding of the physical world. while at the other end of the cline any use is an imposition of social construal on the labeled situation. I not only acknowledge my membership in that category. It seems unhelpful. I am likewise participating in the creation and maintenance of those categorizations. the power of labeling and description is quite limited. it seems less evident that I can do much that matters by such activity. but to what extent is that an imposition. but I cannot be trying to perform the described . when I describe someone else's speech act (she asked me when I had come home). my mention of the label will not change that presence or absence. when closely examined. I will be likely to aect only my own categorization system. At any rate. what am I creating or maintaining? Presumably my human cognitive categories. in this view. I may be engaging in an act of construal. It is otherwise with femininity. in the case of my categorization of something as a cat or a (non-)cat. can be taken for granted as independent of our ongoing construal. although where it stops is unclear. This presumably goes beyond the usual discussions of gender and social-group identity. Crucially. as a category by recognizing it and using it (thinking I am using it depictively). Under most everyday circumstances. so far from our collective bootstrapping mechanisms of embodiment. it seems obvious that one can causally aect that identity by one's own actions. even performative use of category labels is ``only'' a debate about labeling. to abandon the original distinction between performative and depictive. dress. it will appear that at one end of the cline. such categories have a dierent kind of causal character from categories which are more directly based in perception and bodily experience. or cathood. Taking categories from both ends of the cline. If I sustain bookhood. or at most labeling conventions rather than people's actual understanding of the entity in question. of other ``depictive'' uses: little of the social world.Blended spaces and performativity 327 The same might be said. or otherwise behave in a way appropriate to my society's ideas of feminine gender. And when I describe myself or other people according to socially accepted gender categories. as opposed to a recognition of agreed-on categories? Any linguist would agree that uncontested human categorization is maintained and strengthened by use of those categories. but that still seems dierent from causal action to bring such structure into being. But even in the more heavily constructed cases. or even social construction. but participate in the creation and maintenance of that membership. Whenever I speak. or ethnic identity: these concepts are so cognitively constructed. that being categorized as an A or a B is essential to ``being'' an A or a B.
even orders. lesbian. or to make it the case that it was performed. and give commands as straight speakers areÐand also as competent to represent such speech acts. Austin and Searle would doubtless say that this is beyond their immediate purview. rather than remaining silent or silenced (Sedgwick 1993. requires no special conditions beyond very general ones. and thus to make performative uses of descriptions ``take'' in ways which would not have been authorized without social recognition of the categories involved. Particular formal institutional authority may be built up to con®rm such membership. but a claiming of social authority to participate in shaping institutional facts. Chen (1999) suggests that Austin was wrong: a christening carried out by an unauthorized person need not automatically be a simple failureÐit could instead be a disputed case. Again. Describing real or imagined states of aairs (past successful hunts). understood by some people as having carried out a christening and by others as having failed. Gender theorists in particular have also used performativity to refer to the power to perform speech acts. and indeed that it has nothing to say to .. gay performativity apparently refers to the gay community's empowerment to speak of and for itself. with attendant ceremonies. Gay performativity. this would not have been Searle's or Austin's usage of the word. as I could via I ask you when you came home. Sweetser speech act. they are prevented from engaging in certain kinds of important performative activities: US law does not recognize the possibility of marriage for same-sex couples. although it surely involves conceptual construal and social presentation of those descriptions. husband.328 E. is not the same as using descriptions to try to causally in¯uence the future hunt's success. ask questions. which most speech act theorists appear to take as somehow ``given'' by a shared social framework. Social categories are often labeled. therefore. is the problematic nature of institutional or social facts. Livia and Hall 1997).g. bisexual and transsexual English speakers are as linguistically competent to make statements. is not a special kind of performativity. There is certainly nothing in their understanding of performativity which would be special or inaccessible to the gay and lesbian communities: we know that gay. The real issue which gender theory exposes. which are likely to make declarative use of statements using the labels in question (You are now husband and wife). Inasmuch as some groups' chosen categories are not institutionally canonized in this way. questions. But Austin and Searle are assuming that performance of speech acts such as statements. and labels are often used to confer and negotiate membership in social categories (e. marriage. both for performative and depictive purposes. wife). for Austin and Searle's understanding of speech acts. Same-sex marriage ceremonies oer the same possibility of disputed construal.
There is a general understanding that words may represent falsely as well as truly. but even Joe went to Cairo could turn out to be inaccurate. The gender theorists' insight seems to be thatÐlike speech actsÐinstitutional facts themselves exist only by our representation of them: act as if they are altered (or enact them dierently). I would like to take a moment here to consider the special status of nonlinguistic representation. Received Revision received January 2001 University of California at Berkeley . Ritual blends are cognitively strengthened. actions show the reality of someone's intentions and behavior. Cultural and personal dierences in beliefs about causation will likewise in¯uence the performative use of mental spaces: why would one take communion. unless one believed in its ecacy? In looking at the special status of language in this web of mappings. while words may not. Cultural dierences in socially granted causal authorities may dier and cause corresponding dierences in performative powers. 7. need not represent real ``originals''. Conclusions A complex web of mappings shape our uses of the cognitive relationship between represented and representing spaces. but it seems to me that we are inclined to think that they do. actions speak louder than words. and in some sense they really are. and to the similarities between the ways in which linguistic and nonlinguistic forms of representation can be used both depictively and performatively. or to priests or judges to make two people legally married by stating that it is so. Central to this web of mappings is the issue of the causal relationship between a representation and what it represents. Mental space analysis of some of these social phenomena seems like a fruitful ®eld of endeavor. of course. as they say: that is. for example. are we as likely to wonder about the writer's ``original'' as we might do in looking at an illustration of the same sentence? And as for enactment. as in the special authority granted to umpires to make a ball an out by stating that it is so. Pictures. and why it is so important in ritual. The sentence Joe may go to Cairo is speci®cally marked as not representing reality. not only by multimodal involvement of the participants.Blended spaces and performativity 329 their work: they always said that declaratives were dependent on institutional facts. In reading a sentence about a boat in a novel. As all past research agrees. all speakers can causally bring about their own speech acts by describing them. linguists and philosophers of language have paid less attention to the web as a whole. but by the added realism gained from that involvement. well.
1989). Sparrowhawk pays for the boat by magically curing the cataracts of the old man who owns it. 1983. Most notably Austin (1962) and Searle (1969. a couple can become legally married in common law if they represent themselves as married. 120. pp. but only that it will automatically succeed in communicating the illocutionary force of an order. Cf. for example. 5. 3) for an insightful discussion of the aspectual issue here. 58. despite the use of the term performativity to describe this situation. but they are also not considered as agentive or voluntary performance by individual agents. LeGuin . NY]. 94. Though nearly all analysts have problems with these labels: see Durkheim's (1915) discussion. The oddity is thatÐcontrary to the usual interpretation of simple presents of perfective verbs in EnglishÐit is not restricted to this iterative/habitual reading. Thanks to Marco Casonato for drawing my attention to this area. 1993. So. (See A Wizard of Earthsea by Ursula K. p. there is no real distinction between performativity and depiction in this understanding of social structure. and who is generally obliged to appear whenever and wherever he is mentioned. Modern ®ction writers also make use of this convention. in a mutually interdependent context. stating explicitly (Grice 1975) that the Maxim of Quality is really a sub-case of the kind of norm which makes us expect to be handed sugar (and not salt in a sugar-bowl) when we ask for sugar. Sedgwick 1993) 12. if one of them introduces the other to acquaintances as my husband. the old man suggests that he name the boat Lookfar and thus give it metaphorical ``vision'' in repayment for the old man's restored literal vision. could also be a description of a habitual or iterative pattern of promising. Mac Cana (1970). Diana Wynne Jones has a series of novels involving a wizard whose name (or title) is Chrestomanci. who insisted very strongly on the unity between speech acts and other kinds of action. 10. 2. And indeed. as I understand it. 3.) 8. Sweetser Notes 1. in later interactions. The Seabury Press. 11. 6. or my wife.330 E. 4. 1979. 7. (Butler 1990. 13. the boat sometimes seems to be magically aware of the hero's intentions. as well as of weather. a divorce may be necessary to part them subsequently. The Episcopalian communion service. he ends up arguing for the use of the term religion for shared communal ritual enterprises which set out to aect the world. I will touch only very lightly on other senses of the word. And indeed. should someone choose to treat this seriously. Note that of course Searle does not mean that the order will automatically succeed in inducing the perlocutionary eect of compliance. and not merely to attest to its existence. All social roles are considered as performance. In some states of the US. See Langacker (1991: chap. (He has an impressively elegant collection of dressing gowns so as not to be caught at too much of a disadvantage when summoned at less opportune moments. contains such central phrases as the quasi-directives. and magic for individual ones. the State performs via us the roles which it coercively bequeaths to us.) 9. said to communicants as the bread and wine are presented to them. It is unclear to me whether critical theory would answer this question with Neither or Both. The Body of our Lord Jesus Christ keep you in everlasting life and The Blood of our Lord Jesus Christ keep you in everlasting life (The Book of Common Prayer [1977. 365) which clearly indicate that the bread and wine are intended to causally bring about a particular spiritual situation. Rather. . Here I follow closely on Grice.
L. The Church eventually accepted the ordination of women as a result. 133±188. Balthasar 1997 Spatial operations in deixis. Chen. Raymond 1967 Tikopia Ritual and Belief. Bickel. 1993 Bodies that Matter. e. Oxford. Gilles 1985 Mental Spaces. 15. 1962 How to Do Things With Words. Oracles and Magic among the Azande.] 1997 Mappings in Thought and Language.g. References Austin. performed by duly ordained (male) bishops who had the Anglican Church's authority to ordain priests. J.] .. In Nuyts. in Frazer's (1890) The Golden Bough. Note that the ®rst women Episcopal priests to be ordained. 1998 Conceptual integration networks. Stanford. 283±303.). 2000 Grammar and social practice: on the role of ``culture'' in linguistic relativity. 1937 Witchcraft. Allen and Unwin. E.-C. Conceptual Structure.Blended spaces and performativity 331 14. In the sense of Scott Liddell ( and generally in his work). Frazer. 1999 Austin meets the split subject: Alienation and performativity. S. In Niemeier. Cambridge: Cambridge University Press. Judith 1990 Gender Trouble: Feminism and the Subversion of Identity. I would like to thank Agata Kochanska for bringing these parallels to my attention. 16. James 1890 The Golden Bough. Swain). to appear The way we think. Cognitive Science 22: 2. Melinda Y. Evans-Pritchard.). held at the Institut Â Paris-VII. cognition and culture: where to orient oneself in Belhare. Boston: Beacon Press. 113±130. J. Butler. Paper presented at the May 1999 Paris-Berkeley Linguistics Workshop. Cognitive Linguistics 11. [Reprinted Cambridge: Cambridge University Press. Evidence for Linguistic Relativity. Firth. Adele (ed. Fauconnier. and R. but the initial ocial response was that these ordinations were ``valid but irregular''.). Fauconnier. E. 161±192. [New York: Crown Publishers. I should thank Jesper Sùrenson for his input into earlier stages of this paper. Cambridge MA: Harvard University Press. 2000 Compression and global insight. as well as for letting me have input into his ongoing work. in the 1970s. New York: Routledge. Cambridge University Press. W. d'Etudes Anglophones of the Universite Durkheim. Discourse and Language. Cambridge MA: MIT Press. were ordained in a ritually correct ceremonyÐall the right linguistic and ritual forms. 17. Also called sympathetic magic. Emile 1915 The Elementary Forms of Religious Life (trans. New York: Routledge. CA: CSLI Publications. In Goldberg. Dirven (eds. 18. 46±83. Language and Conceptualization. Jan and Eric Pederson (eds. Gilles and Mark Turner 1996 Blending as a central process of grammar. Amsterdam: John Benjamins.
Tambiah. Queerly Phrased: Language. Chicago: The University of Chicago Press. Searle. Cambridge MA: Harvard University Press. 1985 Culture.). Gender. 3: Speech Acts. 1979 Expression and Meaning: Studies in the Theory of Speech Acts. P. 535±558. 1954 The two towers. 283±314. Linguistics and Philosophy 12.). LeGuin. In Cole. Mauss. 1968 A Wizard of Earthsea. Marcel 1972 A General Theory of Magic. and conceptual shifts. Sedgwick. (Original published 1962 as La penseÂe sauvage.aau. Eve Kosofsky 1993 Queer performativity: Henry James's The Art of the Novel.) Liddell. 1983 Intentionality: An Essay in the Philosophy of Mind. Anna and Kira Hall 1997 It's a girl: Bringing performativity back to linguistics. Thought and Social Action. Bronislaw 1965 Coral Gardens and Their Magic. Malinowski. Ursula K. gestures. Bloomington: University of Indiana Press.dk/semiotics/docs/urbino97/Soerensen. Stanley J. vol. T. Peter and Jerry Morgan (eds. http://www. Oxford: Oxford University Press. Jesper 1997 Iconicity in magic: Metaphor and similarity in ritual performance.) Searle. Cambridge: Cambridge University Press. John and Daniel Vanderveken 1985 Foundations of Illocutionary Logic. Cambridge: Cambridge University Press 1989 How performatives work. Cognitive Linguistics 9: 3. trans. Anna and Kira Hall (eds. Cambridge: Cambridge University Press. Ronald W. Image and Symbol: The Cognitive Basis of Grammar. List of papers from international conference on Image in Language and Nonlanguage. . Sùrensen. 1991 Concept. Sweetser Grice.htm 2000 Magic among the Trobrianders: Conceptual mapping in magical rituals. Scott 1998 Grounded blends. 41±58. In Livia. Cambridge: Cambridge University Press. London: Routledge and Kegan Paul. London: George Allan and Unwin Ltd. R. Robert Brain.hum. 1975 Logic and conversation. Claude 1966 The Savage Mind. Tolkien. 1±16. Chicago: University of Chicago Press. and Sexuality. Proinsias 1970 Celtic Mythology. G. Mac Cana. Berlin: Mouton de Gruyter. New York: Academic Press. and Mark Johnson 1980 Metaphors we live by. Langacker. 3±20. Livia. London and New York: Hamlyn. (Original published 1950. Syntax and Semantics. John 1969 Speech Acts: An Essay in the Philosophy of Language. H. Levi-Strauss. GLQ 1. Almen Semiotik. R. Ace paperbacks.332 E. Lako.
Journal of Metaphor and Symbolic Activity 10(3). 183±204. Mark and Gilles Fauconnier 1995 Conceptual integration and formal expression. Ithaca: Cornell University Press. Victor 1967 The Forest of Symbols: Aspects of Ndembu Ritual. .Blended spaces and performativity 333 Turner. Turner.
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue reading from where you left off, or restart the preview.