http://www.beaugrande.com/text_linguistics_at_the_millennium.

htm

Text linguistics at the millennium: Corpus data and missing links

ROBERT DE BEAUGRANDE

Abstract Text linguistics seems to have originated chiefly in order to expand the search for constraints, which was being noticeably impeded by the self-imposed restrictions in a linguistics centred on isolated, invented sentences and abstract formalisms. Yet early attempts to bring the text into the scope of su ch a linguistics now seem inverted: for us the actual text, not the invented sentence, must be the essential linguistic unit, and is sustained by internal systemic organisation and by its external systemic organisation within one or more intertexts. In the coming millennium, this prospect can now finally be documented and clarified by working with very large corpora of authentic texts, whereby we can hope to uncover some of the vital and delicate missing links between language and text. A. Language and text in modern linguistics 1. Modern general linguistics has been a singular enterprise, at times most sharply distinguished from other approaches to language by its self-imposed restrictions. Since its outset, it has been influenced by Saussures (1966 [1916]: 232) aspiration that the true and unique object of linguistics should be language studied in and for itself. In effect, this vision of linguistic science has been pursuing the question: what would language look like when its off by itself and not being used (cf. 14, 82)? 2. An unwelcome answer would be that it no longer looks like a real language. If so, the term language used by this linguistics loses its ordinary meaning, namely: a mode of communication used among the members of a human community. Some common uses in this meaning can be seen in these authentic data supplied in July 1994 from the Bank of English at Birmingham University, the worlds largest computerised text-data corpus, then containing over 200 million words (cf. 60, 64, 89): (1) you should be pleased that the French language has been spared (2) he has no qualifications in teaching English as a Foreign Language (3) its old-fashioned, and its in a foreign language. People are frightened of it (4) I was told afterward that my language was most entertaining (5) fatally damaged? I dont want to use language of that sort (6) there is a lot of bad language and gratuitously oafish behaviour 1

(7) General Kryuchkov used the language of the Cold War when he accused the US (8) He violently opposes the new language law, which makes major concessions to ethnic minorities Sample (1) concerns the French language spoken as a native language by a whole nation, whereas sample (2) concerns the English language as a subject-matter to be taught to, and learned by, people who speak a different native language. Sample (3) implies that a foreign language may contribute to frightening people. Samples (4), (5), and (6) indicate how particular uses of language get evaluated. In sample (7), language covers both the style and the content of what a Russian General said belligerent in tune with the Cold War. And sample (8) mentions a government regulation concerning which language or languages should be used and when, for instance as an official language in a multilingual country. So all these uses relate to real people who either might use a language for communication, or else might be hindered in doing so because they had no qualifications (2), or because the language was foreign (3) or was restricted by a language law (8), and so on. None of the uses matches language in and for itself in the austere theoretical meaning of Saussurian linguistics. 3. This mismatch might explain why Saussure (1966 [1916]: 9, 11) asserted that speech cannot be studied, for we cannot discover its unity; it is only a heterogeneous mass of accessory and accidental facts ( 10, 21f, 39, 82). Later, Chomsky (1965: 4, 201) asserted in similar vein that the observed use of language surely cannot constitute the subject-matter of linguistics, if this is to be a serious discipline; much of the actual speech observed consists of fragments and deviant expressions of a variety of sorts. As if in parallel, these same linguists asserted that the concrete entities of language are not directly accessible (Saussure, 1966 [1916]: 110); and that knowledge of the language, like most facts of interest and importance, is neither presented for direct observation nor extractable from data by inductive procedures of any known sort (Chomsky, 1965: 18). 4. The mismatch is thus a signal that language, being the true subject-matter of linguistics as a serious discipline, is not what real people hear or see in actual speech. Another well-known claim then falls into place: linguistic theory is concerned primarily with an ideal speaker-listener, in a completely homogeneous speech-community, who knows its language perfectly (Chomsky, 1965: 3). Neither this speaker nor this community exist in the real world, as Chomsky (1977: 172) has calmly conceded. So the true nature of language is known only to academics whose degrees in theoretical linguistics somehow equip them with privileged access to perfect knowledge (cf. 29, 34, 83). 5. One of Saussures (1966: 8) most candid acknowledgements also falls into place: whereas other sciences work with objects that are given in advance, in linguistics it is the viewpoint that creates the object. What he did not acknowledge but has been extensively displayed in the subsequent history of modern linguistics is that multiple viewpoints create multiple objects, whereby the meaning of the term language has grown steadily more unstable and obscure. Linguistics has been fragmented into 2

disputatious factions, each assigning to language its own idealised meaning. And much time and print has been expended upon disputing over whose idealisation is better without turning for adjudication to the evidence of real language in actual speech (Beaugrande, 1997a, 1998a, 1998b). 6. What then about the status and meaning of the term text? In the field proposed by linguists from Saussure up through Chomsky, the text would seem to merit no home at all. As Halliday (1994: xxii) has remarked, Saussures understanding of the relationship between the system of language and its instantiation in acts of speaking implied that the text can be dispensed with; and linguistics, for much of the twentieth century, has accordingly been obsessed with the system at the expense of text 7. A few linguists did proffer a home for the text in their theories, but failed to secure it in their own practices of investigation. For Hjelmslev (1969 [1943]: 12), the linguistic investigator is given the as yet unanalysed text in its undivided and absolute integrity (but cf. 88). So linguistic theory starts from the text as its datum and attempts to show the way to a self-consistent and exhaustive description of it through an analysis (1969: 21). The theory must also indicate how any other text of the same premised nature can be understood in the same way by furnishing us with tools that can be used on any text (1969: 16). Obviously, it would be humanly impossible to work through all existing texts, and futile as well, if the theory must also cover texts as yet unrealised (1969: 17) (cf. 88). Even so, Hjelmslev counselled that linguistic theory should seek to describe and predict any conceivable or theoretically possible texts in any language whatsoever (1969: 16f). He even stipulated that linguistic theory should foresee a language without a text constructed in that language (1969: 39f) (cf. 37). 8. Hjelmslevs line of argument implied a programmatic move to situate the text inside the same abstract theoretical sphere as language (or langue) defined by Saussure, whose programme Hjelmslev is known to have heartily approved. We might sense a remarkable expansiveness in the aspiration for linguistic theory to provide one single method of description and analysis: first for the one unanalysed given text, then for all theoretically possible texts, and finally even for the non-existent texts of a language without a text. How this aspiration could be put into practice is obscure: in the four published volumes of Hjelmslevs writings, I could not find even one demonstration. His only directive was that the text is regarded as a class analysed into components, then these components as classes analysed into components, and so on until the analysis is exhausted (1969: 12f). So the text would be like a big chunk of language waiting to be taken to pieces which would then be taken into smaller pieces, over and over, until we arrives at the opposite end from its undivided integrity ( 7, 57). 9. We might contrast Hjelmslevs programme with that of J.R. Firth, who declared that the text is the main concern of the linguist (1968 [1952-59]: 24, 90, 173). All texts in modern spoken languages are considered to carry the implication of utterance and are referred to typical participants in some generalised context of situation (1957 [1951]: 220, 226). So the attested language text should be duly 3

recorded and abstracted from the matrix of experience (1968: 199f, 99). Since it may not be possible or desirable to present the whole of the materials collected during the observation period, some corpus is essential (1968: 32) (cf. sections E, F, and G). 10. Still, Firth did not intend to countermand Hjelmslevs programme by situating the text on the side of speech in opposition to Saussures language, because Firth (1968: 28, 41, 127, 139f) roundly rejected the whole dichotomy between langue and parole. By referring the text to typical participants and generalised contexts and by abstracting it from the matrix of experience, the linguist would lift the text up from the merely heterogeneous, accessory, and accidental plane where we saw Saussure situating speech ( 4). Unfortunately, Firths own four published volumes give only a few sketchy demonstrations, far short in practice of what his writings on theory projected. 11. At all events, the text was likely to remain on the margins of linguistics as long as it was eclipsed by the sentence. Curiously, the sentence did not start out as the prototype of language in the meaning of Saussure. He had argued that although the sentence is the ideal type of syntagm, it belongs to speaking [parole], not to language [langue] (1966 [1916]: 124). However, his reservations distinctly placed the sentence on both sides: in the syntagm there is no clear-cut boundary between the language fact, which is a sign of collective usage, and the fact that belongs to speaking and depends on individual freedom; in a great number of instances it is hard to classify a combination of units because both forces have combined in producing it, and they have combined in indeterminate proportions (1966: 124f; see Beaugrande, 1999a for discussion). 12. These early reservations did not prevent the sentence from later occupying the centre of modern linguistics, most famously when the generative approach defined a language to be an infinite of set sentences (Chomsky, 1957: 13) (cf. 32). Officially, the sentence was a unit purely restricted to syntax or grammar, two terms now used as if they were interchangeable, although they cannot be, as we shall see ( 17f). Yet this restriction could not be sustained during attempts to actually formulate a generative grammar, even for small samplings of invented English sentences. The term sentence was also being unofficially used (among other things) for a semantic unit that should be called a proposition and for a pragmatic unit that should be called a speech act, as if to paper over the indeterminate proportions in Saussures reservation (Beaugrande, 1997a). So just when generative linguists were volubly announcing that the study of competence abstracts away from the whole question of performance and why speakers say what they say, how language is used in various social groups, how it is used in communication, etc. (Dresher and Hornstein, 1976: 328), the concept of the sentence was being quietly stretched and bent in the direction of performance (cf. 50). 13. As long as the sentence was getting stretched, the need to recognise the text might not have seemed urgent. But the restrictions of the single sentence eventually had to become onerous for linguistic 4

investigations even in the narrowed area of syntax or grammar. And some linguists would eventually respond by looking beyond the sentence and at the text. Not surprisingly, much early work in text linguistics emblematically aspired to build upon what now came to be called, for purposes of contrast, sentence linguistics a term that would have seemed oddly redundant to the generative approach. By exploiting the indeterminate proportions implicit in the sentence ( 11), such work need not settle the question of whether the text might be a unit of the actual speech rejected by the linguists who had followed Saussure and Chomsky (cf. 3, 10, 23). 14. The most straightforward strategy for getting the text inside a linguistics designed for the sentence would be to define the text as a sequence of sentences. As such, the text could smoothly inherit the established properties of the sentence, such as being grammatical, well-formed, and rule-governed. Yet disputes arose in the early 1970s over the question of whether those properties were sufficient, so that established sentence linguistics could account for whatever might be found (Dascal and Margalit, 1974); or whether new properties would be found that belong or apply only to texts (van Dijk, 1972). 15. Text linguists naturally favoured the second option, which justified their work as a accredited field. But in retrospect, I find the question inverted. Following such leads as Hartmann (1968, 1971) and Schmidt (1973), we can accept the text as the essential linguistic unit and then explore the status of the sentence as one of the potential segments within a text, and thus one that could explicitly benefit from the previously implicit bending in the direction of performance ( 12, Beaugrande, 1999a). Moreover, the sentence could richly inherit the properties that have since been established for the text, such as being cohesive, coherent, intentional, acceptable, informative, and situational (Beaugrande and Dressler, 1981) (cf. 59). 16. In the process, we might reach the unsettling conclusion that syntax doesnt exist in natural language insofar as the term means a system of formal rules for arranging words together in sentences (cf. 14) (compare Garca, 1979; Givn. 1979). As already implied by Saussures reservation about the sentence ( 11), speakers and writers certainly select and combine words in response to important non-syntactic factors, such as lexical preferences, communicative situations, and personal motivations ( 98-101). To assert that these factors are all beyond the sentence or outside the sentence is to ignore their potent effects inside the sentence and to block off significant resources for description and analysis. 17. The restrictions of syntax become most obtrusive when it gets directly equated with grammar ( 12). Despite the once fashionable notion of a deep structure underlying surface structure (e.g. Chomsky, 1965), a grammar which looks only at the order of words inside isolated sentences and not at why speakers say what they say, how language is used in various social groups, how it is used in communication ( 12) must remain a shallow and superficial enterprise. In English unlike many other languages word-order is sufficiently frozen in some areas to sustain carefully selected Aspects 5

of the Theory of Syntax; and these have indeed been the concerns of generative linguistics until it withdrew into universals and mental representations, where the sentence is no longer crucial (see Beaugrande, 1998a for discussion). 18. A realistic and empirically justified grammar would rather be the front end for all the relevant motivations that speakers or writers typically apply or reflect when they put classes of words (e.g. Nouns) or of word-parts (e.g. Suffixes) in one order rather than another (cf. 101). This lesson might also be drawn from the recalcitrant obstacles encountered by the early projects of text linguistics to construct a text grammar from the top down in the narrow and abstract generative sense. The text grammar for a fairly short text by Bertolt Brecht (Herrn K.s Lieblingstier), which even shows a simple vocabulary and some parallelisms in its phrasing, got tangled in an explosive complexity of rules1 (cf. van Dijk, Ihwe, Rieser, and Petfi, 1972), and the project was eventually abandoned with no official conclusion (cf. 38). The same fate can be predicted for any project to describe texts on comparable levels of abstraction and formality: what gets abstracted away during the formalisation would be vitally needed in order to account just for the order of words, and far more for the choices of words (cf. 40). 19. The converse approach for a text linguistics, and the one that has gradually won out, is to work from the bottom up. We need to examine a comprehensive range and variety of authentic texts and explore what sorts of properties deserve to be accounted for, including, but not restricted to, those of grammar in the broad sense of 18. We can apply whichever categories and concepts of previous linguistics seem productive, but we can also apply ones from adjacent fields, such as literary studies, cognitive science, artificial intelligence, ethnography, economics, and political science (cf. Beaugrande, 1980, 1997a, in preparation) whatever bases we can enlist in exploring how speakers and do select and combine words inside phrases, clauses, sentences, or any other relevant units, such as paragraphs, essays, or science textbooks. 20. This approach has a distinctly Firthian flavour, although we can exploit fields, methods, and resources that were not available to Firth and his pupils. Today, we are far better positioned to observe and collect a corpus of attested language texts and to determine what is typical and generalisable ( 9f). Admittedly, the sheer size of the task remains daunting enough to indicate why Saussurian linguistics was so eager to rule out the study of speech ( 3). Yet the task deserves to be at the top of our agenda for the next milennium. B. Virtual system and actual system 21. As a provisional strategy, some text linguists (myself included) have for a number of years been proposing to view the relation between a language and a text as one between a virtual system and an

actual system. One pioneer for this view was the eminent linguist and language philosopher Peter Hartmann,2 who, together with his pupils (e.g. Siegfried J. Schmidt, Roland Harweg, Walter A. Koch, Gtz Wienold), were the most thoughtful originators of text linguistics. Against any linguistics that might view texts being like Saussures speech, i.e., heterogeneous, accidental, and devoid of unity ( 3), we maintain that the text is internally systemic on its own terms and is externally systemic in respect to other texts ( 47, 99). The text is thus an intersystemic event during which multiple systems interact and converge ( 89). 22. This viewpoint does not mean that the interplay of langue and parole somehow vanishes on the level of texts a consequence inferred from my work by Lindemann (1981: 126). Much in the spirit of Firth ( 10), text linguistics firmly rejects the dichotomy of langue and parole for having been deeply misconceived all along. The major flaw, not widely recognised, has lain in attributing to language an ideal order, and to speech an accidental disorder (Beaugrande, 1998a, 1998b, 1999a). The bizarre implication would be that using a language in speech triggers an abrupt catastrophic transition from stable and integrative order over to unstable and disintegrative disorder. Conversely, the linguist whose study abstracts away from how language is used in communication ( 12) miraculously restores the pristine, ideal order. The practices of linguistic investigation would consist chiefly of idealising language; yet we should be wary of projecting an ideal that is fully disconnected from real language (Beaugrande, 1998a, 1998b) ( 97). 23. Nor again does this same viewpoint mean that texts seem to be exclusively created by an actualisation of lower-level virtualities, and that texts do not dispose of any virtual systemic aspects of their own further consequences inferred by Lindemann (1981: 126). Of course texts hinge crucially upon the virtual systemic aspects; our problem is that virtuality is not readily isolated from actuality when we work with real texts. Lindemann himself proposed that actually occurring texts may be virtualised by exploring their systemic aspects, yet this is precisely what text linguists do in practice though without describing our practices by that term. However, we cannot be just putting virtuality back into a space where it has been actualised out, because texts always retain some aspects of their virtuality as long as the language is still known and used.3 So the text is not just a unit of actual speech (parole, langage, performance etc.) in the senses of Saussure and Chomsky (cf. 3, 10, 13), but is rather a unit which actually links language to speech. 24. Some powerful evidence for the virtuality would be that a single text can be received and interpreted (actualised) in more than one way by different participants in a text-event or even by the same participant at different times. Such is indeed a constitutive principle of literature; the literary text might, as Wellek and Warren (1956: 152) have done, be compared to langue, and each individual realisation to parole. Yet far from navigating between the order of langue and the disorder of parole 7

( 22), the audience recreates a novel order for the textual world. Your motivation a reward for reading or listening to a literary text is thus to be a privileged participant each time in constructing its world, whose order is in part your own creative achievement (Beaugrande, 1988). The same process would apply, on a far less conscious plane of creative awareness, for the actualisation and re-actualisation of any text; literature merely accentuates and thematises the process to deepen and broaden our understanding of the human situation. 25. The implication would be that the same text need not have the same order for everybody. At least in its fine details, each actualisation is unique and unrepeatable. Moreover, each participant has a partially individualised knowledge of the virtual system, with greater variation in the lexicon and lesser variation in the grammar ( 83ff). So we must resolutely address the questions of how and how far the respective actualisations of any text can agree or coincide well enough for a text receiver to feel confident of understanding what the producer meant; or for several participants to receive a text in what is thought to be the same way Evidently, the degree of convergence among participants in a textevent is achieved on line through the interaction of multiple systems whose design both anticipates and adjusts to these achievements. 26. The production and reception of a text would be complementary transitions between a more open mode and a more closed mode of systemic order; and these two modes would determine each other in a continual dialectic. Contrary to the assumptions of linguistics cited above, we would conclude that these two modes of order must be quite proximate: the order of language is more closed and the order of texts more open than has been widely acknowledged before ( 56). 27. Text linguistics accordingly needs to develop models of how the one mode of order sustains and tunes the other. We might postulate a cycle whereby the language actualises to sustain and tune the texts, whilst the texts virtualise to sustain and tune the language. My intransitive usages of the two verbs may sound esoteric, since the agents of these reciprocal activities are of course the participants in the text-event; but the activities are normally just by-products of communicative interaction and do not match what the participants are consciously doing or intending to do (Hartmann, 1963). The exceptions would of course be the poets, whose enterprise consists of seeking and testing new ways of organising language (Beaugrande, 1978, 1988). 28. Our own enterprise as text linguists consists of consciously virtualising texts whenever we actualise them for the purpose of drawing inferences about the order of the virtual language system ( 23). A key question, still to be adequately explored, is our own version of the well-known problem of participants versus observers (cf. Harweg, 1968). How can far our conscious and attentional actualising-for-virtualising correspond to, or accurately describe, the unconscious and automatic actualising and virtualising of ordinary text producers and receivers; and what might occur during our 8

processes of making conscious and focusing attention? Quite plausibly, the potential of the texts being investigated gets considerably more elaborated precisely insofar as we prolong and intensify the cycle postulated in 27 (see also 83ff). 29. In addition, the principle of each participant having a partially individualised knowledge of the virtual system ( 25) must apply to us too. What might be our status as individuals and within the wider community? Even if we could set aside our academic training and consider ourselves reasonably typical, we could not yet consider ourselves representative, due to several constraints. The most obvious constraint is that the language experience of any individual, even one who has read or written extensively, is only a small part of the experience of the community. Instead of conveniently taking the community to be homogeneous and purporting to have privileged access to the perfect knowledge of the language ( 4), we need to lay open the degrees of real uniformity and diversity among language users and among their texts to large-scale adjudication with substantive evidence. 30. A less obvious constraint on being representative could be inferred from the virtuality I have suggested that texts always retain ( 23). This cannot be directly measured by their manifested uniformity, because virtuality is by definition open and dynamic, and what is manifested is the actuality that partly confirms it and partly specifies or modifies it to suit the context. Instead of vowing that knowledge of the language is never presented for direct observation nor extractable from data ( 3), we can assume that knowledge of the (virtual) language never totally converges with any set of actualities, but does move steadily closer toward convergence as the set gets larger and more diversified. In that sense, we all text linguists or not remain language learners throughout our lives as we accumulate experience with texts. Text linguists are just exceptionally self-conscious learners, observing and extracting whatever we can in the knowledge that there is always far more beyond it all. 31. Still another constraint on being representative is the unresolved uncertainty about intuition providing the source for a grammar to describe the intrinsic competence of the idealised native speaker (Chomsky, 1965: 24). Contrary to such well-known claims, unaided intuition is not broader and deeper than language experience but narrower and shallower. It is best secured for the more stable and general frozen areas of the virtual system whose virtuality seems static and hence independent of actual speech, whence the preference for restricting linguistics to a few Aspects of the Theory of Syntax (cf. 17). But intuition is not well secured for the dynamic cycles of actualising and virtualising precisely because of the continual tuning. The interacting systems settle down in precise detail only at the respective stages of the interaction itself. So the native speakers intuition may well not be reliable for telling you what they would say until there arises a real-life occasion when they do say it. Intuition operates most smoothly after the fact in making sense of what has already been said ( 99).

32. This overall line of argument also indicates that the language system never settles down in a synchronic dimension; consequently, all attempts since Saussure to construct a synchronic description are doomed to be partial, restricted, and provisional. However, these limits should not disband the synchronic linguists so much as rather resituate their enterprise. Like total convergence ( 30), synchronicity is a factor we can move closer to yet never attain. It is not set apart in some ideal language disconnected from speech but represents the totality of simultaneous actualisings of the virtual system (cf. Harweg, 1968: 142). So we cannot remotely achieve any synchronic description, however partial, by dispensing with texts (cf. 6); we need bigger and better means for examining what large sets of texts have in common. 33. We thus return to a deep question: if language and text, or virtual and actual, are closely interconnected yet never converge, how then are their connections sustained? On a highly general plane, we might stipulate: the language is a set of texts; each text is one member of the set. We might interpret this stipulation by viewing a language as one vast continuous text, as the totality of what has been written and spoken (or perhaps even thought) in a language (Harweg, 1968: 142, my translation). 34. We might feel reminded here of the definition (already cited in 12) of a language being an infinite of set sentences (Chomsky, 1957: 13).4 The term infinite merits closer examination, as I have shown elsewhere (Beaugrande, 1998a). The infinity was hypothesised on the simplistic mechanical basis of recursive devices (Chomsky, 1957: 23f) whereby any sentence could, in theory, be made longer or more complex. In practice, the set of infinitely long or complex sentences is an empty set and thus an intractable object for constructing a grammar. Nor can it plausibly be attributed even to the perfect knowledge of ideal speakers, who also form an empty set of humans (or superhumans) (cf. 4). 35. Still, for the sake of argument, we could infer that if a text is defined to be a sequence of sentences ( 14), then generative theory would allow for an infinitely long text. That text would be theoretically equivalent to one superlong sentence with sentence boundaries as sentential connectives (Fodor and Katz, 1964: 491). Yet such a text would also inescapably belong to an empty set. At most, our theorising could plausibly postulate the concept of an infinitely long intertext, whose practical correlate could be a conversation in which the last word can never be said. Such a notion has a post-modern or poststructuralist flavour, though we are arguably still dealing with an empty set. What is given is a real set of intertexts such as conversations, each of which is at any one moment finite, though quite possibly still open, such as the discourse of history, which could be definitively terminated only if history came to an end. 36. Halliday (1997: 6) has for some time preferred to reverse the principle and characterise a language as an infinite system generating a finite body of text; but following a recent discussion where I pointed out that, in terms of mathematical theory, the infinite must include combinations with infinitesimal 10

probabilities, he now favours replacing infinite with indefinitely large. The finiteness of the set of texts in existence at any given moment in time is hardly disputable, but is a weak constraint for a living language, especially one spoken as widely as English. Another weak constraint comes from the indefinite largeness of the set of potential texts in English. The key question, as pointed out elsewhere by Halliday (1993), remains: how does the set of existing texts constrain the set of potential texts such that, at any given moment, some ways of adding to the existing set are substantially more probable than others? 37. Problems with sets also arose when, influenced by the high-level aspirations of generative linguistics, some text linguistics essayed to formulate the constraints for distinguishing the set of texts from the set of non-texts (see Dressler, 1972 for discussion). At the time, these constraints seemed intuitively plausible in theory, but we can now see why they proved intractable in practice. The attempt to create a non-text is always just one step within an ongoing textual event, and thus still a sub-text, however unconventional. Its textuality is locally disrupted but globally sustained. So in both theory and practice, the set of non-texts is one more empty set and therefore obeys no constraints. Nor can it be identified with the set of non-existent texts for which Hjelmslev would have us postulate a language ( 7f), since those would still be texts, although virtual texts of a language whose textual process, i.e., its actualisation, is itself purely virtual (cf. Hjelmslev, 1969 [1943]: 40). 38. I shall cite just one more constraint on our being representative: the practices of text linguistics who analyse and describe texts inevitably add to the set of texts. We thus incur the additional challenge of probing the relations between the set of texts we are analysing and the set of texts we are producing as we go along (cf. 83ff). This challenge corresponds to Firths (1968) distinction between language under description and language of description, and to his epigram about linguistics being language turned back on itself. Evidently because science in general has preferred to regard its own texts as a purer code, eschewing rhetoric and simply reporting natural fact (Bazerman, 1988: 6, 14), this challenge is still far from fulfilled. My own studies have consistently found that the relation between scientific texts and their domain they purport to account for is complex and unstable, especially when that domain is language itself, as in modern linguistics. Far too little attention has been devoted to the practical strategies of a linguist producing and receiving a text about language, even though they have significant effects upon the operations and results of theorising (Beaugrande, 1984, 1991). 39. For the present discussion, one effect I would highlight has been the widespread textual strategy among linguists of inventing sets of artificial pseudo-texts, usually isolated sentences which are either trivial in their grammaticalness (like the man hit the ball) or else wildly bizarre in their ungrammaticalness (like *ball the hit ball man the). These are intended to capture some constraints, optimistically called rules, of the ideal system being called language; and their artificial status may

11

even be deemed to purge them from the heterogeneous and accidental qualities of speech (cf. 3). But in the account I have proposed here, the production of pseudo-data would tune the language system out of its natural frequencies, and thus lead to the postulating of a pseudo-system which does not deserve to be called a language. 40. A second effect I would highlight has been the widespread textual strategy among linguists of formalising texts, apparently to render them more amenable to a scientific account (as in Koch, 1971; Ballmer, 1975). The actual result is to create a set of semi-texts that are artificial in a different way from pseudo-texts, namely in having some features or aspects of texts whilst many other features have been abstracted away ( 12, 18). These semi-texts are not trivial; in practice, they may be almost incomprehensible to everyone but their creators. The production of semi-data would thus also tune the language system out of its natural frequencies, and lead to the postulating of a semi-system which deserves to be called a language no more than does a pseudo-system. The practices of the linguists during this production remain quite arbitrary and ad-hoc until we have a full and creditable account of which features or aspects should or should be abstracted away and how the remainder should be formalised. 41. The practices of formalising texts might correspond to a theory of the text itself being a theoretical unit underlying a practical unit called discourse (cf. van Dijk, 1972). Here, we might recall Hjelmslevs aspiration to get the text inside the same abstract theoretical sphere as Saussures language, which Hjelmslev himself never attempted to demonstrate in practice ( 8). And, as I have remarked ( 18), the demonstrations for text grammars displayed an explosive complexity of rules and features, precisely insofar as a text is not solely a theoretical or virtual unit but also a practical and actual unit. If we disregard or abstract away from its practicality and actuality, we lose a host of significant constraints; trying to replace or reconstruct them all with rules and features is a gratuitous, self-defeating exercise. 42. Instead of complicating the relation between language and text with sets of pseudo-data and semidata, the preferable alternative strategy would be to let the texts represent themselves ( 82). The texts would retain the same representation as recorded speech or writing within the experience of the community of participants in text-events. Special annotations such as representing the intonation of speech in a written format should be clearly authorised by the purposes of the investigation and should not impede unduly comprehension or access to the data. 43. The practices of text linguists would then not entail transforming texts into some other representation but rather transforming our own modes of accessing and exploiting texts as data sources. We need to base our own authority not upon holding higher academic degrees in theoretical linguistics (cf. 4, 29) but upon examining large sets of authentic textual data (section H). 12

C. Theory and practice 44. These deliberations about the theories and practices of text linguists counsel us to move onto a higher plane (doing what Peter Hartmann used to call berhhung).5 There, we could define a language to be a general theory of human knowledge and experience evolving in a dialectical relation to texts as a set of practices for working out the theory (cf. Hartmann, 1963; Halliday, 1987). By that definition, all the members of a language community are implicated in the theory and practice of language and text (cf. Beaugrande, 1997b, 1998c). More precisely, they sustain and share a dynamic theory which evolves through a criss-crossing interaction of many implicit and partial theories about how the practices of everyday life and ordinary talk are organised. Whilst the diverse contexts come and go, different partial sub-theories (rather than the whole theory) are applied and specified in and through the practices. 45. To be sure, a language is a highly unique type of theory. It cannot be effectively tested and verified or falsified in the familiar manner of a scientific theory, because it partially constitutes what it postulates. We cannot get outside language in order to talk about it without implicating ourselves in it (cf. 38). Such would seem to be the aspiration of projects to formalise language, but they merely end up replacing it a semi-system whose relation to real language presents even more knotty problems ( 40). 46. Nor again can we say that any one language is a more correct theory or a more valid theory than any other. The potential of any language for expressing human experience is infinite in theory, though always finite at any one moment of practice. Some languages have, for historical of political motives, had their potential more fully actualised in specific domains, such as science and technology. Such is true of English but by no means validates the language to be the superior one so often extolled in the discourse of International English (discussion in Pennycook, 1994; Beaugrande, 1999b). 47. Now, by the definitions proposed above, what has been officially called a theory of language or a linguistic theory so far would be a meta-theory, whereas the texts we produce to formulate and expound the theory would display our meta-practices. According to my line of argument here, these meta-domains would be fundamentally different depending on what their object of investigation is declared to be. In the meanings of Saussurian and Chomskyan linguistics, language studied in and for itself is a theory about itself, about pure theoreticalness disconnected from the practices of actual speech. In consequence, their linguistic theory would be a meta-meta-theory, as shown most programmatically in the English title of Hjelmslevs Prolegomena to a Theory of Language. This theory is doubly remote from practice, and the practices of doing linguistics in general and building linguistic theories in particular are radically under-constrained. Not surprisingly, many books on theoretical linguistics have been sharply critical of the prior state of the field and have proposed to make a fresh start (Beaugrande, 1991: 334ff). Over time, linguistic theory appears to offer a stunning variety and disparity of clashing doctrines, and striking divergences in terms, slogans, and technical contrivances 13

(Jakobson, 1970: 12). 48. But if our object is declared to be authentic texts, then a text-linguistic theory stays proximate to the practices of real language-users, and our own meta-practices would not be radically under-constrained. Indeed, we can productively invest our own status as participants in analysing or interpreting data produced for a community to which we rightly belong, such as the readers of the Times, so that our textlinguistic practices are reconciled with the practices in our object domain, though of course not simply converging with them. Against the tendency of scientific discourse to hide itself, as if it were somehow not writing at all (Bazerman, 1988: 14) ( 38), a science of text would gain by displaying and even analysing its own discourse and the latters implication in the constitutive and constructive practices of text-events (see section H). 49. Working back toward ourselves from the opposite side, we should also devote far more attention to the theoreticalness of producing and receiving texts in and about everyday life. Most producers and receivers would probably consider themselves ordinary practitioners or practical users of a language, who have no interest or authority concerning theoretical matters. They might be irritated or perplexed to be told that knowing a language endows them with one among the most extensive and powerful theories a human can achieve, far beyond the revered but restricted theories of science. But until they can acknowledge this vast endowment they will not be adequately empowered to use this resource for gaining access to knowledge and society (cf. Beaugrande, 1997a, 1998c). 50. In sum, a chief reciprocal task for a science of texts would be to cultivate an active sense of the practicalness of scientific texts (ours in particular) and the theoreticalness of ordinary texts, within a programme for reconciling the two sides within a dialectical interaction. Doing so would require reconstructing the missing links between theory and practice, and between a language and the texts in that language. I shall now consider some prospects as we look ahead to the new millennium. D. Missing links 51. Perhaps the term missing links sounds overly dramatic. After all, the linkage between a language and a text between theory and practice is achieved whenever human beings communicate, although they dont recognise their achievement in those terms. But in terms of scientific investigation, our theories and models of this achievement undeniably manifest some missing links. I have been suggesting here that the toughest problems have stemmed from the assumption that a language has a quite different mode of organisation than does actual speech (or texts), and from the corresponding aspiration to describe language independently of actual speech. So the urgent move now would be to develop a more text-like view of a language and a more language-like view of a text ( 55). 52. One promising approach would be the systemic functional linguistics prominently developed by

14

Michael Halliday, who was once a pupil of J.R. Firth. Halliday was among the first to realise that the order of a language must be systemic in modes that anticipate, reflect, and support the systemic uses of the language in texts (cf. 21). Firths (1968: 192) vision of a text as a longer piece to be described as a relational network of structures and systems at clearly distinguished but congruent levels, converging again in renewal of connection with experience accorded with Hallidays (1994: xiv) vision of a language as networks of interlocking options. Quite plausibly, the networks constitute or contribute one type of missing links whose precise nature is now beginning to be explained. 53. As Halliday (1997: 6) has recently acknowledged, his approach must contend with vast exponential size of the networks needed to represent systemic potential as alternative combinations of features. The network needs to present not just those sets of options that are currently being instantiated but also the open-ended further expansion of that potential (1997: 7). Moreover, any such network representation should reflect the cline between options which are highly probable and options which are highly improbable ( 36) (Halliday, 1993). Some options might be fully possible but rarely or never get used ( 62). 54. So the networks for a language would be both too unwieldy and too under-determined to serve directly as networks for a text. In exchange, the networks for texts, such those I introduced for the pedestrian rocket text (Beaugrande, 1980), are too compact and specific to represent the networks of the English language. Once more, we seem to detect some missing links, this time between two types of systemic networks. 55. The strategic place to look for missing links would now be in a very large corpus of authentic texts. Such a corpus can offer our best means for promote both a text-like view of language, and a languagelike view of texts ( 51). It can also display the competence of the language community bent very far toward performance, as well as the communitys performance bent very far toward its competence (cf. 12). We might undertake to complement the systemic functional approach by exploring appropriately sorted corpus data which may provide evidence for our system networks, allowing them to extend much further in delicacy while continuing to model language as potential (cf. Halliday, 1997: 24). 56. Corpus data are so eminently suited to informing us about networks because they offer concrete displays of the constraints upon how sets of choices can interact. In the lexicon part of the lexicogrammar of English, these constraints constitute the collocability in the virtual system, and the textual actualisations are the lexical collocations. In the grammar part of the lexicogrammar, these constraints constitute the colligability in the virtual system, and the textual actualisations are the grammatical colligations. Following Halliday (1961) and Hasan (1986), we can say that the lexical choices are more delicate, and the grammatical choices are less delicate; ultimately, the lexicon would be the most delicate grammar, whilst the grammar would be the least delicate lexicon. Every point can 15

access every other. Thus, what is collocable can constrain not just the collocations but the colligations; what is colligable can constrain not just the colligations but the collocations; and so on (cf. 63). Such constraints ensure that the order of language is reasonably closed, whilst the order of texts is reasonably open ( 26). 57. These prospects counsel us to adopt a synthetic approach of exploring how choices converge in texts, rather than the analytic approach of subdividing the text into steadily smaller pieces, as counselled by Hjelmslev (who never showed us how) ( 8). We would then highlight the systemic continuity of the text event: not just the mere co-presence of smallest meaningful units (e.g. morphemes), but the interaction of meanings within a coherence that is not just the sum of the meanings of the parts (Beaugrande, 1984). 58. Perhaps an analogy from particle physics might be helpful. There, the four forces have recently been reinterpreted not as brute pushes and pulls, but as information exchanges of messenger particles: photons for the electromagnetic force, gluons for the strong force holding the nucleus of the atom together, bosons for the weak force regulating radioactive decay, and presumed on grounds of consistency but not yet observed gravitons for gravity. By analogy, we could envision semantons as sub-symbolic messenger particles of meaning being interchanged among the virtual meanings of morphemes or whole words (cf. Smolensky, 1989; Beaugrande, 1997a). Through these information interchanges, some word-choices appear to attract each other at varying degrees of strength, in analogy to magnetism or gravity, and thus to determine which areas of virtual meaning are being actualised to constitute the context (cf. 63). 59. The interactive constraints and attractions indicated in Fig. 1 would continually supply and sustain the links missing links insofar as they have eluded our investigations to date between the language and the text. Part of the actualised output would be the cohesion and coherence of the text as standards of textuality, but also the other standards identified in text linguistics: intentionality, acceptability, informativity, and situationality (Beaugrande and Dressler, 1981) ( 15). The most missing links of all begin to emerge, perhaps like the cold dark matter astronomers believe to be holding galaxies together, namely the links of intertextuality. For the first time ever, we can assemble and compare the evidence for detailed and delicate intertextual constraints shared among hundreds or even thousands of related collocations or colligations. Instead of seeing the text as a set or sequence of sentences, we can finally see it as a contribution to intertext ( 35, 87).

16

Text linguistics at the millennium 2 E. If I were you... 60. For a programmatic illustration, I shall survey some corpus data kindly provided to me by Stephen Bullon, Publishing Manager at COBUILD (Collins Birmingham University International Language Database) in late 1998 and early 1999 from the Bank of English at Birmingham University, the worlds largest computerised data corpus, then containing over 329 million words of running text (cf. 2) 61. I wanted to explore the Conditional colligations if I were and if I was, because they have been items of dispute among teachers of English grammar. My sampling consisted some 800 occurrences out of the total of over 4000 a partial sample of course, but a week of hard work to sort out all the same. In respect to colligability, I found a strong attraction for the Second Person Pronoun6 you, and the data indicated why. The coherence of the data was centred on delivering advice (9), a warning (10), or a threat (11), whilst the cohesion was centred on the two exophoric Actors I and you: (9) Id get some sleep if I were you. Youll need to be up at six to catch the early morning flight from

Heathrow. (10) The builder looked at it and said, I hope youre not thinking of filling that thing with water. I wouldnt if I were you itll go through the floor. (11) The blood of the mob is up! If I were you, Id clear out of town now with as much as you can carry The speakers intentionality was to imply: Im not telling you what to do, Im just telling you what I would do from my own standpoint if I happened to be in your place. Some leeway is reserved for the acceptability of the hearer who can decline without either side losing face. The informativity is low when the situationality alerts the hearer about what might or should be done. But in some contexts, the advice was clearly improbable and informative, e.g.:

17

(12) a girl asked what she could do to stop her boyfriend ejaculating prematurely. If I were you, I would drop him straight away, came the answer (13) Im thinking of sending my paper to the Journal. Is there room in the next issue? Benedict turned to him. If I were you, he said carefully, but with thickened accent, I would shred it. Huntley bridled. I beg your pardon? When estimating the probabilities of data, we apply our own knowledge of intertextuality being actualised on demand. Someone leaving from harrowing Heathrow Airport had better get plenty of sleep (9) in order to survive the ordeal; and someone who has stirred the blood of the mob had better clear out of town (11). But some girl who seeks advice from an agony aunt about her boyfriend does not expect to be ordered to drop him straight away (12); and someone who has written a paper would bridle at being told to shred it (13). 62. Other pronouns allowed by the cohesion of English were found to be disfavoured by colligability. Whereas if I were you appeared in 282 data samples and if I was you in 37, others colligations were at best marginal, some hovering between 10 and 20 and some close to or equal to zero. The totals were: if I was he if I was him if I were he if I were him 0 17 3 18 if I was she if I was her if I were she if I were her 2 5 0 6 if I was they if I was them if I were they if I were them 0 11 1 10

These data indicate which of the virtual colligations of English cohesion are or are not probable in their colligability for actualisation ( 53). Interestingly, the probabilities were about even for whether the Verb was in the Indicative was or in the (presumably) Subjunctive were, even though both colligated with the Object Pronoun much more readily than with the Subject Pronoun. Yet when the Pronoun was you, which does not differentiate between Object and Subject, the old Subjunctive still proved by far the more probable choice. Perhaps this choice is felt to signal a Contrafactual Modality and thus to be more tentative and polite for purposes of face-saving; or perhaps the whole colligation is simply chosen and produced as one frozen unit (compare theory and practice in 98). 63. Several constraints on colligability and collocability appeared when the Conjunction if had a meaning like whether. This meaning colligated with the Indicative was possibly to leave open the question of whether something is Contrafactual and resisted colligating to the right with a Pronoun. A common collocation to the left was a Verb expressing uncertainty, like ask, wonder, or not know, as in (14-16). (14) One day Mrs Luppin remarked that I was looking a little off-colour and asked if I was feeling all 18

right. I told her about my sickness. (15) then I got pregnant and I started wondering if I was going to be able to do things right (16) I didnt know if I was going to do this assignment [] But late last night, I decided to. In such contexts, the Verb see never meant perceive with the eye, nor did tell ever mean inform, although these are the definitions listed as the most probable by conventional dictionaries. Instead, the meaning was find out or determine as in: (17) I wrote seven poems to enter, very much borrowed from the Philip Larkin style. I borrowed his sense of depression, too, so all my friends were phoning to see if I was having a nervous breakdown. (18) No one with a gun had shown himself above the roofline but how could I tell if I was being watched? Such data show how the attractions among meanings can influence not just the actualisation of one potential option, but also the actualisation of one potential meaning for the superficially same option (cf. 58). The colligation as if colligates in turn with Nouns or Modifiers to the right, and collocates with specified Verbs to the left.

F. Intertextual actualisations of theory and practice

64. I shall now attempt a less usual demonstration with corpus data, examining how the pair of terms theory and practice are actualised in contemporary English texts, and what the implications might be for situating these pairs within a genuine dialectic. My data were again kindly provided from the Bank of English by Stephen Bullon at COBUILD, this time in January 1999. 65. First, we can look at text samples wherein both theory and practice appeared. The data indicated a cline between the pole of easy connection and the pole of uneasy disconnection. One data source for easy connection were book titles on a motley variety of topics, such as: Warfare in the Twentieth Century: Theory and Practice; Life Insurance: Theory and Practice; Islam in Theory and Practice; and The Family Interpreted: Feminist Theory in Clinical Practice. Intriguingly, the idea of warfare or life insurance having their own theory intuitively seems informative enough to suggest that the author intended to upgrade and dignify a heavily practical activity (cf. 93). In return, we might intuitively grant the status of a religion like Islam or a philosophy like feminism being a theory, and see the informativity focused upon how they get carried over into practice against the contrary trends toward a secular or patriarchal global society.

19

66. The other main data source for easy connection was found among services offering both theory and practice in a package deal. Some were merely trendy, like herbal medicine (19), while most were concerned with education, sometimes in specialised fields like accounting (20), but often in general education (21-23), which we shall be seeing again ( 68, 95). (19) a course designed to give students instruction in the theory and practice of the use of herbs for medicinal purposes and herbal preparations to maintain healthy skin and hair. (20) Financial Reporting explores accounting theory and practice, including such topics as the development and objectives of financial reporting (21) The practicum is concerned with relating theory and practice, helping prospective teachers to understand and practise a wide range of teaching skills (22) the adult education course contained five nights of theory and practice, for three hours each night (23) The course of study ensures that theory and practice are closely related, providing an opportunity for students to continue with work These trends can be easily by verified by surfing the Internet. On 14 February 1999, the Alta Vista search engine returned 105,469 home pages for the verbatim collocation theory and practice. Alongside such frankly pedestrian topics as Good Cooking, Horse Racing, Belly Dancing, and Late Victorian Wallpaper, I found courses or surveys in Health Activism, Grading for the Fashion Industry, Radiometric Calibration, Rubber Injection Moulding Common Object Request Broker Architecture, and far more others than I have room to list here. This rampant proliferation indicates how empty the terms have become, conveying a vague promise of being comprehensive and thorough ( 75). At times, the promise sounded distinctly breezy: (24) it also helps the student to make the vital connection between esoteric theory and mundane practice, between the future of the planet and the recycled Pepsi can 67. In still other data, the disconnection between of theory and practice was openly acknowledged, though still in confident hopes of a reconnection. One topic was the economic organisation of a society: (25) As business starts to bloom in Central and Eastern Europe, theory will increasingly turn into practice. (26) The Marxist argument against the separation of theory and practice is well known Marx himself would have vehemently denied that the separation of theory and practice which he strove to overcome could disappear when free-market business starts to bloom in formerly socialist countries. But then he would have undoubtedly affirmed that the collapse of the polity in those same countries resulted from authoritarian refusals to put the theory of socialism into real practice 20

(Beaugrande, 1997a: 397) (cf. 80). 68. Toward the middle of the cline shown in Fig. 2 we could put the data indicating that the connection between theory and practice is not clear (27) or is open to disagreement (28): (27) each group allowed to exercise its rights to freedom and self-determination, but it is not clear what this policy will mean in terms of practical politics (28) Sometimes members of the community disagree over the practical definitions given to values everybody professes in the abstract, such as liberty or equality Further along toward the pole of uneasy disconnection was the prospect of fundamental cracks in society that are actually widening: (29) the fundamental cracks in both society and theory which were laid bare in the 1960s continue to widen; theory becomes increasingly separated from practice and, thus, incapable of addressing those cracks Predictably, the social domain suffering the most conspicuously was education, which I often noticed urgently searching for a reconnection, as in: (30) addressing the challenge of how we can make comprehensive education work for those for whom theory and practice have remained far apart. That is a debate to relish (31) Perhaps the primary benefit of this block was its attempt to link theory with practice. Most teacher preparation programs have a separation (32) a cross section of teachers from within school who can apply their learning in situ, bridging the gap between theory and practice by generating theory from practice and practice from theory (33) the actors within the schools unique culture, provided theory for action fostering a very different attitude towards theory than the usual presentation of disembodied theory which many teachers experience on external in-service courses 69. A cracked society might offer hidden motivations for supporting democratic institutions trade unions in theory whilst destroying them in practice (34); or for manipulating the theory of an institution like the law in order to justify ones perversions in practice (35). (34) In theory, he is a great supporter of trade unions, because, in theory, hes a socialist, but in practice, he has been the most successful destroyer of trade union power in the British print (35) The problem, Mr Olson thinks, is not legal theory but legal practice although lawyers clearly manipulate the former to justify their perversions of the latter. 70. Against my own intuition to approve of connection between theory and practice, some data indicated 21

disapproval: (36) The scope of coercive violence was extended: Pornography is the theory, rape is the practice. Pornography is violence against women. These made pithy slogans (37) Up until that point the proud theory and the practice of the majority was that you lived with your family until you met the man you would marry (38) It was a symbolic and deferential democracy, but the symbols were rooted in substance: myth became the mortar binding theory and practice The connection was designedly tendentious for pornography and rape (36), but merely patriarchal for marriage (37) and mythical for democracy (38). Still, to assert that a myth brings substance to symbols does seem audaciously informative, as does the metaphoric mortar suggesting a solidity and strength that are only imaginary. 71. My personal attention was attracted by the topic of war or warfare, whose theory and practice we saw being invoked in a book title ( 65). Some data implied that war is essentially a set of practices that can be pursued without much interest in theory (39); or that the theory can be invented after the fact (40). (39) he was an independent, practical soldier with a distaste for theory; a fighter who believed that war provided the only lessons of real value to a soldier. (40) He did not convert theory into practice but exemplified instinctive practice and later derived theory from it. Few soldiers have so impressively But most of the data concerned an official theory expressly devised to justify the practices of warfare such that they will not be recognised to grossly contradict all the humane ideals of the society. Ironically, the author of the most celebrated result, the just war theory was a godly philosopher whose sources were eminently theoretical, such as Plantonism (41). Incorporating the Christian position on an intellectual basis can harmonise this theory with the theory of a holy war, God being squarely on your side (42-43). By emphasising intention over reality, a just war can be glibly translated into a just peace, (44); and can project a resounding triumph of good over harm (45). Yet if the theory really requires that every less-violent means be exhausted (46) and that provisions be made for innocent civilians (47), then no war I can think of has fit its practices to the theory certainly none in this century. (41) Augustine expressed a new attitude toward conflict by formulating the just war theory. He adapted rules of warfare developed by classical thinkers such as Plato and Cicero to the Christian position. (42) They forgot how intellectually respectable the Christian theory of holy war once was 22

(43) theres something called a just war theory which is justified an awful lot of wars in the worlds history, where both sides feel they call on God (44) one of the seven points of the just war theory is right intention, and that is sometimes translated as the intention to have a just peace (45) the just-war theory clearly requires that the good to be achieved outweighs the harm that is to be done (46) Essentially, the just-war theory requires, above all, that every less-violent means be exhausted before war can be justified (47) Likewise, the just-war theory requires that there be provisions made for innocent civilians The real but unofficial theory of modern warfare has been described by Tony Wilden (1987: 27), following Brownmiller (1975), in terms where the power of God is equated with limitless violence: The object is to destroy the will to resist; the target is the entire population; the strategy is terror; the means is torture; the usual end is death; most of the victims are women and children: the worst instrument is rape. To do this you simply let your men loose [with] the power of God over anyone and everyone without a weapon or the strength to fight back. According to the just war theory, in contrast, the object would be to secure a just peace and the sovereignty and honour of the nation; the targets are exclusively military installations; the strategy is tactical manoeuvring; the means is manly valour; and the end is the victory of the good side (ours). Obviously, the just war theory can do wonders for upholding morale and for deflecting world-wide outrage over real atrocities, which can be officially deplored as unfortunate lapses during an otherwise just enterprise. 72. This contrast between theories indicates that the official position of governments conducting a war or preparing to do so of course as a means to keep the peace (M. Thatcher) will embrace the just war theory whilst conceding that it is not, or not always, put into practice (48-50), due to unforeseen problems (50). Improvements in technology are cheerfully expected to establish the connection, e.g., to improve the practices of bombs landing two-thirds of a mile off their targets (51). (48) One theme runs throughout, however the dichotomy between the theory of war and its practice (49) As this chapter will endeavour to show, the gap between theory and practice in strategic bombing is wide. (50) after only 30 months of war, the original theory of strategic bombing had been stood on its head because of practical problems in carrying out missions (51) Warden believed, however, that modem munitions offered a precision that would marry theory and practice. A B-17 bomber in World War II had a circular error probable of 3300 feet, meaning that half 23

of the bombs dropped far from their targets 73. Governments have invoked supportive theories to explain why warfare could not be avoided. My data turned up the domino theory (war must be fought here lest neighbouring countries fall to the enemy) (52-53), and the deterrence theory (54) (continual threat of war is needed to deter your enemies from war). Both theories were heavily exploited during the Cold War, especially in Southeast Asia (53), yet the armaments for war were illogically retained after Cold War had ended (54). (52) We were faced with the Cold War and the domino theory," the 60-year-old former national serviceman said (53) He [Nixon] mined harbors, talked about decisive military action to end the war, and used the domino theory to justify the need to stop communism in Vietnam. (54) Ordinary people, including those who subscribed to the deterrence theory during the cold war, are angry that politicians have not taken the opportunity to move towards a nuclear-free world The all-time prize for public mystification was surely earned by the limited war theory (55-58) concocted to justify the lavish preparations for war after nuclear weapons had foreclosed the chances of surviving, let alone winning. This theory was obliged either to ignore the enemys response (57) or to naively imagine the sort of enemy whose thinking is entirely guided by game theory (58). (55) How can conflict be waged between nuclear-armed adversaries without leading to mutual destruction? Limited war theory attempts to provide a framework (56) We had to invent a theory that would allow us to fight on the edges without nuclear technology. This theory is called limited war. Its premise is that we and the Soviets can wage little wars, and that each side will refrain from going nuclear (57) And in the event of a conventional attack the scenario was for the first use of battlefield weapons to keep the nuclear war limited and contained. That was the theory. But there was no account taken of the enemys response (58) Limited war theory had been built on the assumption that the opponent was cautious and valuemaximizing, not fanatically determined The basic premise of this theory which I have never once heard anyone seriously defend is vastly more remote from real practices than anything imagined by Plato or St. Augustine: a superpower with an staggering stockpile of nuclear armaments would refrain from using these even in times of utmost desperation, and tamely accept defeat, humiliation, and dispossession by the rival superpower who, the propaganda machines have insisted, is the very incarnation of evil. If the Soviets rejected such an absurd theory (59), then they were motivated as much by common sense as by escalation theory. 24

(59) bringing in escalation theory from political context is the fundamental motivation behind Soviet rejection of many aspects of limited war theory. This is crucial, because if a war is to be limited in scope through deliberate restraint A far more accurate label would have been substituted war theory, where the little wars (56) between two or more clients of the superpowers allow the latter to stereotype short of the logical end-game of mutual destruction (55). Vietnam was supposed to be one such substituted war, but it didnt go according to the theory of the superpowers for limited war (60), precisely because the enemys response was fanatically determined. (60) Vietnam saw the application of many of these ideas. Whether or not the Americans were consciously applying limited war theory, many of the concepts underlying and shaping their policy were the same as those shaping limited war theory 74. I have focused on the topic of warfare as a conspicuous demonstration of how an official theory can get theorised not to account for the real practices, but to provide a camouflage for the real theory. I suspect many far less conspicuous parallels could be found wherever the real theory that would actually account for the real practices would be inadmissible in a modern democratic society. If so, the apparent global trends toward democracy from the 1960s onward might be grasped as a re-theorising of postwar societies that transformed official theories whilst leaving the real practices largely unchanged (see Beaugrande, 1997a for extensive discussion and references). G. Intertextual actualisations of theoretical and practical 75. The corpus data for samples wherein both theoretical and practical appeared did not reflect the same trends as those with theory and practice in any straightforward way, but parallels could be detected. The collocation theoretical and practical was often enlisted to enhance the commercial or professional value of some essentially practical activity or training by implying that it is comprehensive or thorough (cf. 66): (61) The program will incorporate theoretical and practical training within the schools own commercially-operated hotel, just outside Canberra. (62) the weekend is designed to give Queenslands food and wine professionals and serious amateurs access to the latest information, practical and theoretical. (63) Media Studies, National Certificate. This course gives a basic introduction, both theoretical and practical, in media studies as a starting point for a career in this area (64) The course is both theoretical and practical with the emphasis once again on gaining real experience under the guidance of a senior BBC Radio Trainer.

25

The collocation was also enlisted to suggest the serious nature of ones deliberations: (65) This chapter is concerned with the practical and theoretical implications of seeing child abuse as a matter of mens violence to babies, children, and young people (66) More recently there has been a renewed interest in fundamental social problems such as poverty, and a search, both on the practical and on the theoretical level, for explanations of the persistence of problems despite the rise of institutional welfare. Criminology presented an interesting exception: either remaining reticent about awkward matters that were both theoretical and practical (67) or else shunting them off into ethical matters in order to profess neutrality in both theory and practice (68). (67) has often been ignored or avoided by critical criminologists. Carlen has observed that there is a radical reticence about theoretical and practical questions concerning the desirability, recognition, denial and concomitant control of the power to punish. (68) Most criminologists are engaged in theoretical and practical projects without considering their ethical implications. They believe that they can be ethically neutral This aspiration is quite natural in view of the questionable ethics of punishment (not to mention the endemic beatings, rape, and torture in our prisons) in a society whose official theory is to rehabilitate criminals. 76. Where the data indicated some uneasy disconnection, the theoretical was typically disapproved and the practical was approved: (69) Those experiences were of crucial importance to me, for they removed my evolving approach from a purely theoretical context and re-rooted it in a practical theatrical foundation (70) These are not theoretical niceties, they are fundamental practical moves forward in market-led strategic change (71) Perhaps most distressing is the continued emphasis on theoretical academic work, not practical or technical skills needed for the marketplace. (72) most of the courses echo a cut-the-crap utilitarianism. Background details, historical origins and theoretical models give way to practical understanding-and-doing-skills. Compare this glowing vision of a practical man being nearly superhuman: (73) He was a talented, capable, practical man, articulate, imaginative and resourceful, the sort of person who could be relied upon to cope with any situation Some of the data were less obvious, but a bias toward the practical might still be inferred: 26

(74) But it is fair to say that today, in general, the environments own standing has become more of a matter of theoretical and spiritual interest than a real practical constraint on the bringing of environmental litigation (75) he also spoke of freedom in terms which party critics like Roy Hattersley might endorse as being dependent not just on theoretical choice but on the practical ability to exercise it, the freedom to work and develop talent (76) These medical schools, often controlled by a church hostile to any birth control, enrolled almost no women, and stressed theoretical expertise rather than the practical knowledge embodied in folk medicine. Theoretical interest by itself wont accomplish much to save the environment (74), ensure freedom (75), or improve the control of 'women over their own lives (76). 77. Occasional data recommended a clear decision for the one side or the other: (77) there was nothing for it but to distinguish sharply between the theoretical man, who seeks understanding, and the practical man, who feels compelled to take political action (78) In an age of specialists, the organizational aspects of mathematics function more tidily if people specialize either in the theoretical areas of the subject or its practical ones. Because most people feel happier working in one or the other of these two styles In still other data, connecting theoretical to the practical was advocated, though the prospects might seem dubious: (79) if the lesson of history is that strong links between theoretical research and practical applications are a key factor in scientific progress, the effort to bring the two together is crucial (80) though not led directly by Marxist theory, they partially demonstrated the correctness of that theory, showed that socialism was a practical possibility and not just the theoretical result of idle speculation by intellectuals who had no firm grasp of the realities (81) the method of sociological intervention aims to open social movement participants eyes to the wider theoretical significance of their practical activity and thereby to catalyse the movements potential for creating a new type of society So far, we are left to assume, science has failed to bring together research and applications (79); socialism has been mainly speculation divorced from realities (80); and sociological intervention has brought us no nearer to a new type of society (81). If, as I would surmise, such failures are related to the endemic disconnections of modern democratic society ( 74), expecting a total change seems, well, impractical. 27

78. When either theoretical or practical occurred without the other term, some differences in collocability could be readily noticed. The term theoretical collocated in titles or descriptions with distinctly academic topics, such as mathematics, astronomy, physics, chemistry, biology genetics, physiology, anatomy, physiology, economics, psychotherapy. In contrast, practical collocated with distinctly non-academic topics such as photography, seamanship, suturing, cooking, dining, woodworking, gardening, summer pruning of tree fruit. Equally remote from academic concerns were books offering practical advice like these: (82) this extremely practical book which mixes common-sense advice with many trade secrets gleaned from extensive experience (83) Courtship: Wisdom For Wooers And The Wooed offers practical guidance on condom etiquette, cheating, children, and, toughest of all, how to introduce your other half If, like me, you are puzzled about why introducing your other half should be tougher than rearing children, youll have to buy the book to find out. 79. The uses of theoretical for academic topics were most pronounced when collocated with ostentatious technical terms, e.g.: (84) In technical-theoretical parlance, countries are required to have factor endowments in the same cone of diversification (85) Bio-inorganic chemistry: experimental and theoretical studies of speciation involved in trace element bio-availability, and pharmaceutical trace metal interaction The link between theoretical and authority was intensely emblematic when opposed to rampant empiricism (86) or when traced back to the features of language suitable for sounding expert (87). (86) Rampant empiricism without a firm grounding in theoretical structures will result in multiple research projects with fragmented yield and low efficiency (87) how they are perceived on some authority scale might influence how the audience receives their experimental report or theoretical argument. They know they have to sound expert. However, they havent really figured out which features of language The term theoretical itself was evidently thought to be one such feature of language, witness its actualisations to lend some intellectual respectability to dubious topics: (88) Psychoenergetics in Theoretical and Practical Aspects, in Proceedings of the Third International Congress on Psychotronic Research. (89) Reincarnation Field Studies and Theoretical Issues, in B. Wolman, ed., Handbook of Parapsychology 28

(90) To escape from this state Machen put his theoretical knowledge of occultism to practical use, and after using a process that seems to have been some sort of magic The term was also actualised to make unpleasant realities seem more abstract, such as tolerance of immigrants (91), profiteering by means of non-existent capital (92), and death by cancer (93). Warfare popped up again for a theoretical intervention in Africa (94), where the relief did not include food, water, or medical supplies, but manpower and technology, presumably to relieve the endangered power-holders who safeguard European interests. (91) However, until immigrants began arriving in the mid-1980s, this tolerance was a largely theoretical exercise. (92) the stockmarket boom inflated their theoretical capital to phenomenal levels (93) However, the same assumptions that underpin calculations of theoretical deaths also predict an increased risk of fatal cancer in direct proportion to the increase in body weight. (94) A theoretical taskforce of soldiers, warships, and military aircraft was sent to carry out a relief mission in an imaginary African country 80. The corpus data prominently represented two theoretical fields. Astronomy was a favoured topic because so much of it remains hypothetical, due to the vast distances, masses, energies, and time lapses involved in stars or galaxies: (95) Crawford admits that these ideas are theoretical and highly speculative. I havent a clue how faster-than-light travel or communication might be achieved. (96) This provides a picture of the Multi-verse, as some call this theoretical froth of universes. Because conditions in the black holes are so extreme, even the laws of physics might change On the other hand, socialism was a theoretical topic because humans can put it into practice only if the society equitably redistributed its resources, and the holders of power are grimly determined to achieve just the opposite. The data showed people preferring to view socialism as a very distant objective (97), far withdrawn from the common-sense world (98). In response, the advocates of socialism were found mistrusting the purely theoretical and striving to gain a firm foundation upon the social experiences of the community (99) and to reconcile the theoretical work of intellectuals with the manual labour of proletarians by sharing a practical stance (100). The irony seems consummate if the working class attained a theoretical position that disconcerted the academic labour theorists (101). (97) Despite Ramsay MacDonalds commitment to a Labour Party whose theoretical aim was to transform capitalism into socialism, as leader of the party he saw this as some very distant objective (98) Her [Hannah Arendts] thought is a kind of reflection that deals with essences, with what is most 29

abstract, general, and theoretical and it involves withdrawal from the common-sense world of appearances. (99) political wisdom is not to be found in the theoretical speculations of isolated thinkers but in the historically accumulated social experiences of the community (100) By the same token, the theoretical work of intellectuals was thought to be no less practical than the manual labour of proletarians (101) labour theorists of the period were aware of and disconcerted by the fact that the working class was developing its own theoretical position on the nature of society a position that was antagonistic to the interests of established society 81. As with so many terms, my two here have been appropriated for consumerism, witness the breezily offered theory and practice of wine and hotels (61-62). Collocated directly with a commodity, theoretical appeared just once and in a weird context (102), the intended meaning being intellectual, I would assume. But practical collocated with all sorts of commodities, including shoes, cameras, and tableware along with some whose practicality might well be doubted (103-04). (102) these theoretical disco records blended morbid introspection and exotic rhythms (103) from our brochures showing a wide range of colourful and practical items, the upholstery with practical zip-on covers for the flame-retardant foam cushions (104) Another tip is to look through the peculiar catalogues that drop through the letter box and out of magazines. Practical ideas abound. A portable plastic bidet, for example. Once they are the proud possessors of this ingenious receptacle I would rather not picture the practices of the proud possessors in (104)! H. Into the millennium 82. The corpus data adduced above of course do not cover the entire range of ways in which two pairs of terms theory and practice and theoretical and practical are being actualised in contemporary English texts, but do suggest some significant tendencies worthy of further research. Such data undeniably let us observe what language looks like when it is being used (cf. 1), and to exploit texts by letting them represent themselves ( 42). We confront neither Saussures mass of merely accessory and accidental facts plus Chomskys fragments and deviant expressions, nor again the latters completely homogeneous speech-community ( 3f). The data reflect an ongoing dialectic between the heterogeneity within the speech-community and the homogeneity encouraged by shared participation in text-events. 83. This dialectic embraces us text linguists as well, since we belong, in our private lives, to the 30

community of English speakers being enticed with foam cushions and plastic bidets (103-04) along with books of advice on how to cheat on our spouses (using a portable bidet to purge the evidence?) and how to give introductions (81) so that likely candidates to cheat with might think were single. In our professional lives, we can productively invest our own status as participants in analysing or interpreting textual data produced for the community ( 48). Doing so should help to keep our own texts free of the complacent or obscure abstractions that imply having some privileged access to the perfect knowledge of the ideal speaker ( 4, 29, 34, 38), or purporting to miraculously restore the pristine, ideal order which actual speech had converted into disorder ( 22). 84. Yet our efforts to achieve the viewpoint of Firths typical participants in some generalised context of situation ( 9) must remain continual work in progress. Exploring corpus data is the most effective means I have found for checking and enhancing the typicality of my own English, though I shall never get finished. I have repeatedly noticed areas where my own usage was at best incomplete and at worst plain wrong, but most often just too vague. This finding is quite natural even for a Professor of English, because, like everyone else, I have had only a limited range of real-life occasions for saying things (cf. 31). I can discuss the theory and practice of education and frequently have, but I would be hopelessly lost in discussing the theory and practice of belly dancing or rubber injection moulding ( 66). 85. Throughout my career, I have suspected that most disagreements among linguists stemmed not from their official disputes about theory-building, e.g., whether the semantics should be interpretive or generative, but from their uneven capacities or dispositions for interpreting language data. The unevenness might reflect the conscious or unconscious striving of some linguists to isolate a Saussurian langue or a Chomskyan competence whilst seeming to disappear behind data deliberately invented to be so trivial like the man hit the ball that they seem to interpret themselves. Such linguists look like theyre struggling to forget or cancel the fact that they themselves are speakers and listeners who continually navigate between langue and parole, or between competence and performance and who are therefore not positioned to isolate the one side from the other. Moreover, since ordinary language experience is not mainly the interpretation of trivial isolated sentences, their results are hardly likely to be representative of anyones competence, let alone the whole communitys. 86. My professional situation has been decisively coloured by my own academic history. My work in linguistics began with the translation of poetry (Beaugrande, 1971, 1978), which has trained me to squeeze a great deal of interpretation out of the data put before me, because as a translator I must base my concrete practices upon my own interpretation of the original text. So I was never deeply impressed with the kinds of theoretical linguistics that shied away from practices and retreated either into invented sentences or arid formalisms, though I only gradually came to appreciate why the whole enterprise is fundamentally misconceived, as I have explained here.

31

87. My later work has been concentrated in language education in a broad sense. I have thus become extremely sensitive to variations and uncertainties in and about the usage of English, which I can see creating serious obstacles for prospective learners or teachers of English. Indeed, the inequalities in access to English world-wide are currently a major engine of those social and economic inequalities that our modern democracies officially claim to have overcome (Beaugrande, 1997a, 1999b). So my own interest in corpus data is firmly situated in a larger search for new directions and initiatives in a millennium which will undoubtedly decide who may or may not speak or be heard. 88. Again, I have good reason to mistrust a linguistics proclaiming that speech cannot be studied ( 3). If, as Saussure acknowledged, the viewpoint creates the object, then we have no use for the Chomskyan ideal speaker, who has no viewpoint at all. What text linguists must do is search our data for signs of convergence of viewpoints among the community instead of glibly making our own personal viewpoint supply all the norms and standards. Hjelmslev was misled: the linguistic investigator is not given the as yet unanalysed text ( 7) but the actualised text; and until I re-actualise it, I am given nothing at all. And if (as he vowed) linguistic theory must also cover texts as yet unrealised ( 7), then the theory must equip us for engaging with the texts that are realised. 89. So my enterprise hinges vitally upon gathering extensive data about the was in which different peoples actualisations may converge or diverge. We thus return to the critical dependence of text upon intertext which, I have argued, turns our attention in the opposite direction from seeing the text as a set or sequence of sentences (cf. 14, 35, 59). By examining sets of related choices across large numbers of texts, we can finally rescue intertextuality from the theoretical and practical limbo where the arcane adumbrations of post-structuralism had located it, and can make tractable estimates of what is more or less probable about specified combinations. Similarly, we can finally grasp the text as a genuine intersystemic event internally systemic in the mutual attractions among options, and also externally systemic in the tuning of those probabilities through other texts within the intertext ( 21). These options form the networks that are more open than any one actual text yet more closed than any virtual system of a language ( 27). 90. How far an intertext might extend is, at this stage, totally unclear in theory. In practice, you come to a trade-off between increasing the delicacy of your survey by increasing the quantity of data versus keeping the presentation of your findings from becoming too bulky to be manageable or indeed publishable (cf. Sinclair, 1998: 82). I tend to level out with around one hundred samples culled out of larger data set; for the data cited the present study, I scrutinised some 1600 samples. But we need to form very large teams to examine correspondingly large samplings whose size and range can be decided by practical goals, such as providing a browsers corpus for self-paced learners of English as a Foreign Language (Beaugrande, 1997b, 1998c),

32

91. A comparable trade-off between delicacy and quantity applies to the length of your individual samples. The default length in the Bank of English search programme when I used it in July 1994 was 80 characters, with the key word or words roughly in the centre, but I rapidly found this inadequate. The context rarely settles down in any informative delicacy at less than twice that many, and my samplings for this study, provided by the patient Mr Bullon, were roughly 200 characters, rounded up to begin and end with complete words. Even that much is far from ideal, and my selection of data to present was partly steered by the reliable delicacy of contexts at the specified length. This too is a merely practical and somewhat adventitious artefact, to be overcome when our resources allow. 92. All these factors set the context for composing the present text about theory and practice: in society, in education, in science, in linguistics, and finally in a modest sampling of contemporary texts. I have sought to build a case to support my assertion that human practices, including ones with texts, are necessarily theoretical, but the underlying theories are largely implicit and unrecognised. In return, official theories are constantly being constructed and promulgated to satisfy the aspirations of a society of social group to regard itself as human, fair, equitable, rational, and so on. Evidence to the contrary can be expediently theorised out, especially by classifying it all as individual accidents or mishaps which are (of course) deplorable but which do not discredit an official theory. 93. I have proposed to describe the textual data along a cline ranging from easy connection to uneasy disconnection between theory and practice ( 65). Some data confirmed my intuition of disconnections whose purport would be: in theory it should be like this but in practice its not. Yet some other data implied either that a connection can be taken for granted (just buy this book and youll see), or else that it has not yet been achieved but can be and should be (just enrol in our all-new training course). Often, theory and practice or theoretical and practical seemed to convey little more than a vague and mystified promise to be comprehensive or thorough in dealing with matters as diverse as Horse Racing, Late Victorian Wallpaper, Health Activism, or Grading for the Fashion Industry ( 66). The mystification could be more easily noticed if practice were taken out, e.g., if you proposed to enlighten the world on the theory of belly dancing or on theoretical wine. 94. In general, theory or theoretical collocated with the lexical items for academic topics, whereas practice or practical collocated with those for non-academic topics, especially commodities ( 78f). In respect to attitudes, theoretical was approved for academic contexts of sounding expert and authoritative ( 79), but disapproved in contexts of being abstruse or unrealistic ( 76). In contrast, practical was nearly always was approved, e.g. for a man who was talented, capable, and resourceful (73). 95. The most consistent preoccupations and those with which, for the reasons given here, I felt the greatest resonance were found for the topic of education ( 66, 68). There, a separation (31) or a 33

gap (32) was admitted along with cheerful projects to reconnect: to banish disembodied theory (33) and (more bluntly) to cut the crap (72). The sheer convergence among these data indicate that the prospects are dim: if centuries of efforts have not failed to connect up the theory and practice of education, what can you hope to achieve, say, in five nights (22)? The failures of education cannot even be understood as long as they theorised out as individual mishaps of pupils being lazy, undisciplined, disobedient, and so on. 96. The real (but unofficial) theory of modern education derives from the persistent assumptions that the main objective must be the passive absorption of theoretical knowledge about topics like algebra; and that the proportions of right and wrong answers in test situations are the most valid measures of individual achievement. Such assumptions could explain why conventional education, both in its theories and in its practices, has remained so insulated from the theories and practices of social and professional life in rapidly changing societies. This insulation is usually accepted as a self-explanatory precondition for education even though it instils pervasive doubts about relevance which severely alienate many teachers and learners. No significant improvements can occur whilst education is enlisted in the crucial disconnection of modern democracy to sustain equality in theory and inequality in practice (Apple, 1984; Aronowitz and Giroux, 1986; Beaugrande, 1997a). We will need a full-scale curriculum transformation for a better linking of theory and practice to improve everyones learning, by combining study with work, by a more democratic restructuring of schools, and closer links with their communities (van Rensburg, 1994: 130). 97. The double-tracking of our societies and their institutions has been faithfully reflected in a modern linguistics that theorises out of language the practical realities of human interaction. The resulting idealisations, misleadingly still called language, are not just empirically vacuous and scientifically invalid, but also socially irresponsible insofar as they have denied to society the insights needed to promote equality and democracy in and through language. Here too, a sweeping transformation is urgently needed to bring real language back into the centre; and, for the reasons I have sought to expound, we can reliably do so only by surveying very large corpora of actualised texts. 98. The analytic or descriptive tools of linguistics can be productively enlisted as theoretical guidelines for practical goals. I can only pick out some salient points to illustrate here. The word-order theory and practice is quite frozen. The reversed order practice and theory appeared on a mere 8 home-pages of the Internet (compared to the 105,469 noted in 66), and all but one of these came from a single source, namely the 1st International Conference on the Practice and Theory of Automated Timetabling in Edinburgh in 1995. Either the normal word-order is simply treated as one invariable unit; or the first item is also the one ranked higher than the second (compare Cooper and Ross, 1975, on world order). The second explanation does not fit the attitudes I have noted above. Nor is it plausible in view of the far 34

more comparable home-page proportions of 28,743 for theoretical and practical and 8361 for practical and theoretical. Either way, these two sets of home-pages were almost all academic or educational, often on austere topics, such as optimal input filter design or uncertainty and complexity in automated knowledge acquisition. Nothing so mundane as horse racing or belly dancing. 99. The probabilities of local colligations and collocations were better defined in the corpus data, which are propagated under different conditions from Internet home-pages. At one end, the probability proved high for the fixed collocation theory and practice to collocate with lexical options from the topic of education, such as course (21-22, 33, 63-64, 71), teacher (21, 31-33), and student (19, 23-24). At the other end were the extremely improbable combinations like theoretical disco records (102) and theoretical froth of universes (96), or practical guidance on condom etiquette (83) and practical portable plastic bidet (104). Still, neither end of the probability scale could be captured in term of rules for the grammaticalness or well-formedness of sentences, all of which would be at inappropriate degrees of abstraction (cf. 18, 41); yet our competence is not overly taxed by interpreting the data in context, thanks to the systemic nature of texts (21). Here too, I would conclude that intuition operates most smoothly after the fact ( 31). 100. John Sinclair (1991: 495), a grand pioneer of corpus linguistics, has remarked that the variations are much more interesting than the regularities. My own interest has been most piqued by the active dialectic between variation and regularity, or in Saussurian terms, between collective usage and individual freedom ( 11). The influence of collocability and colligability often applies not to some specific choice (such as course) but to some category of choices, such as the set of commodities that are advertised to be practical because, like shoes or cameras ( 81), the practicality of some brands might be doubted. Even ostentatiously arcane combinations like trace element bio-availability would fall under the terms of bio-inorganic chemistry (85), and the intentionality may well may be to convey more scientific authority than, say, how new species are formed when trace elements are available. The acceptability of readers is compelled, though they might well prefer, say, to read not (86) but (86a): (86) Rampant empiricism without a firm grounding in theoretical structures will result in multiple research projects with fragmented yield and low efficiency (86a) Empiricism lacking a theoretical basis will lead to fragmented and inefficient research projects The high degrees of informativity in science texts are often merely apparent and masked behind obscure modes of expression (cf. Bazerman, 1988). 101. Word-order might be iconic for the packaging of commodities like the flame-retardant foam cushions (103) which do burn but not so fast as others and are displayed here like rubbery fire-wardens; or like Reincarnation Field Studies (89) which slyly packages imagination with observation. Or,

35

parallelism in word-order can mark a contrast, e.g., between the theoretical man, who seeks understanding, and the practical man, who feels compelled to take political action (77); or a contrast that one aspires to resolve, e.g., between the theoretical work of intellectuals versus the manual labour of proletarians (100). Or again, word order can regulate informativity by guiding focus, e.g., toward a
FACtor KEY

in scientific PROGress for which our effort is CRUcial (79); or by evading the problems of

Agency through a Passive when political wisdom is found (by whom?) in the historically accumulated social experiences of the community (99). Such data show some of the many ways whereby grammar can as the front end for textual and intertextual motives in choosing and arranging words and word-classes ( 18). 102. Such then are some prospects I see for text linguistics as we head into for the next millennium, ready or not. Now that we have access to very large corpora of authentic texts, the linguistic issues I have aired and numerous related ones can be reassessed from the bottom-up. Some major problems, though far from solved, are finally being recognised and placed on the agenda, pending the solutions that will require extensive explorations of data. Moreover, the textual aspects of social and cultural issues can be expansively explored, such as education and science. The delicate links between language and text many of them still missing from our established theory and models are also the most precise regulators of nearly all significant human theories and practices. And, because language partially constitutes what it postulates ( 45), and because the actualisation of a text participates in constructing a world ( 25), the linkage will do much to decide the future evolution of societies already caught up in swift and epochal transformations.

Notes 1. This problem of complexity was also aired in Dascal and Margalits (1974: 85) critical view of text grammars, but their solution was simply to stay with sentence grammar. 2. Hartmanns (1963: 91) own term for the virtual was potential. In another volume, Hartmann (1964:

51, my translation) expressed concern lest the pair of terms render the main term system unduly vague, and he proposed instead the pair system for the potential and complex for the actual. But for us the term complex must seem too overloaded with other associations, especially due to the recent prominence of complexity theory. 3. Harwegs proposal (following K.L. Pikes unified theory) to recognise the emic text alongside the

etic text points in a similar direction, except that emic was defined to be language-internal and etic to be language-external (1968: 152, my translation). I doubt whether this distinction can be sustained in a comprehensive description of authentic texts.

36

4.

In this early passage, the wording was finite or infinite but in Chomskys subsequent formulations

we consistently find infinite. 5. I recall Hartmann using the term in our discussions, though I could not find it in his landmark

volume Theorie der Sprachwissenschaft (1963). 6. Grammatical terms are written in upper case throughout for easy recognition.

References

Apple, M. (1985). Education and Power. Boston: ARK. Aronowitz, S., and Giroux, H. (1983). Education under Siege: The Conservative, Liberal, and Radical Debate over Schooling. South Hadley, MA: Bergin & Garvey. Ballmer, Thomas. (1975). Sprachrekonstruktionssysteme. Kronberg: Scriptor. Bazerman, C. (1988). Shaping Written Knowledge: The Genre and Activity of the Experimental Article in Science. Madison: University of Wisconsin Press. Beaugrande, R. de. (1971). Der Einfluss Baudelaires und seiner Nachfolger auf die Dichtung Stefan Georges. Berlin: Free University MA dissertation. Beaugrande, R. de. (1978). Factors in a Theory of Poetic Translating. Amsterdam: Rodopi. Beaugrande, R. de (1980). Text, Discourse, and Process. Norwood, NJ: Ablex. Beaugrande, R. de (1980). Linguistics as discourse: A case study from semantics. WORD 35: 15-57. Beaugrande, R. de (1988). Critical Discourse: A Survey of Contemporary Literary Theorists. Norwood, NJ.: Ablex. Beaugrande, R. de (1997a). New Foundations for a Science of Text and Discourse. Stamford, CT: Ablex. Beaugrande, R. de (1997b). Theory and practice in applied linguistics: Disconnection, conflict, or dialectic? Applied Linguistics 18: 279-313. Beaugrande, R. de (1998a). Performative speech acts in linguistic theory: The rationality of Noam Chomsky. Journal of Pragmatics 29: 1-39. Beaugrande, R. de (1998b). Linguistics, sociolinguistics, and corpus linguistics: Ideal language versus real language. Journal of Sociolinguistics 3: 128-139. Beaugrande, R. de (1998c). Society, education, linguistics, and language: Inclusion and exclusion in theory and practice. Linguistics and Education 9: 99-158. 37

Beaugrande, R. de (1999a). Sentence first, verdict afterwards: On the long career of the sentence. Word 50. Beaugrande, R. de (1999b). Theory versus practice in language planning and in the discourse of language planning World Englishes. Beaugrande, R. de (in preparation). A New Introduction to the Study of Text and Discourse. Beaugrande, R. de, and Dressler, W.U. (1981). Introduction to Text Linguistics. London: Longman. Brownmiller, S. (1975). Against Our Will. New York: Simon and Schuster. Chomsky, N. (1957). Syntactic Structures. The Hague: Mouton. Chomsky, N. (1965). Aspects of the Theory of Syntax. Cambridge: M.I.T. Press. Chomsky, N. (1977). Language and Responsibility. New York: Pantheon. Cooper, W., and Ross, J.R. (1975). World order. In Papers of the Parasession on Functionalism, R. Grossman, J. San, and T. Vance (eds.), 63-111. Chicago: Chicago Linguistic Society. Dascal, M., and Margalit, A. (1974). Text grammars: a critical view. In Probleme und Perspektiven der neueren textgrammatischen Forschung 1, J. Ihwe, H. Rieser, and J. Petfi (eds.), 81-120. Hamburg: Buske. Dijk, T. van (1972). Some Aspects of Text Grammars. The Hague: Mouton. Dijk, T. van, Ihwe, J., Rieser, H, and Petfi, J. (1972). Zur Bestimmung narrativer Strukturen auf der Grundlage von Textgrammatiken. Hamburg: Buske. Dresher, E., and Hornstein, N. (1976). On some supposed contributions of artificial intelligence to the scientific study of language. Cognition 4: 321-398. Dressler, W.U. (1972). Einfhrung in die Textlinguistik. Tbingen: Niemeyer. Firth, J.R. (1957). Papers in Linguistics 1934-1951. London: Oxford. Firth, J.R. (1968). Selected Papers of J.R. Firth 1952-1959 (ed. Frank R. Palmer). London: Longmans. Fodor, J. and Katz, J. (1964). The structure of a semantic theory. In The Structure of Language, J. Fodor and J. Katz (eds.), 479-518. Englewood Cliffs: Prentice-Hall. Garca, E. 1979. Discourse without syntax. In Syntax and Semantics 12, T. Givn (ed.), 23-49. New York: Academic. Givn, T. 1979. From discourse without syntax: Grammar as a processing strategy. In Syntax and Semantics 12, T. Givn (ed.), 81-112. New York: Academic Halliday, M.A.K. (1961). Categories of a theory of grammar. Word 17: 241-92. 38

Halliday, M.A.K. (1987). Language and the order of nature. In The Linguistics of Writing, N. Fabb, D. Attridge, A. Durant, and C. MacCabe (eds), 135-154. Manchester: Manchester University Press. Halliday, M.A.K. (1993). Quantitative studies and probabilities in grammar. In Data, Description, Discourse, M. Hoey (ed.), 1-25. London: Harper Collins. Halliday, M.A.K. (1994). Introduction to Functional Grammar. Second Revised Edition. London: Arnold. Halliday, M.A.K. (1997). Linguistics as metaphor. In Reconnecting Language, A.-M. SimonVandenbergen, K. Davidse, and D. Nol (eds.), 3-27. Amsterdam: Benjamins. Hartmann, P. (1963). Theorie der Sprachwissenschaft. Assen: van Gorcum. Hartmann, P. (1964). Syntax und Bedeutung. Assen: Van Gorcum. Hartmann, P. (1968). Zum Begriff des sprachlichen Zeichens. Zeitschrift fr Phonetik, Sprachwissenschaft und Kommunikationsforschung 21: 205-222. Hartmann, P. (1971 [1968]). Texte als linguistisches Objekt. In Beitrge zur Textlinguistik, W.D. Stempel (ed.), 8-29. Munich: Fink Harweg, R. (1968). Pronomina und Textkonstitution. Munich: Fink. Hasan, R. (1987). The grammarians dream: Lexis as most delicate grammar. In New Developments in Systemic Linguistics, M.A.K. Halliday and R. Fawcett (eds.), 184-211. London: Pinter. Hjelmslev, L. (1969 [1943]). Prolegomena to a Theory of Language. Madison: University of Wisconsin Press. Jakobson, R. (1970). Main Trends in the Science of Language. New York: Harper and Row. Koch, W.A. (1971). Taxologie des Englischen. Munich: Fink. Lindemann, B. (1981). Review of R. de Beaugrande, Text, Discourse, and Process. Journal of Literary Semantics 10/2: 120 - 127. Pennycook, A. (1994). The Cultural Politics of English as an International Language. London: Longman. Rensburg, P. van. 1994. Education with production in South Africa: A proposal for a seminar. Journal of the Foundation for Education with Production 10/2: 127-133. Schmidt, S.J. 1973. Texttheorie. Munich: Fink. Sinclair, J.McH. (1991). Shared knowledge. In Georgetown University Round Table on Languages and Linguistics 1991, J. Alatis (ed.), 489-500. Washington, DC: Georgetown University Press. 39

Sinclair, J.McH. (1998). Large corpus research and foreign language teaching. In Language Policy and Language Education in Emerging Nations: Focus on Slovenia and Croatia, R. de Beaugrande, M. Grosman, and B. Seidlhofer (eds.), 79-86. Stamford, CT: Ablex. Smolensky, P. (1986). Information processing in dynamical systems: Foundations of harmony theory. In Parallel Distributed Processing: Explorations in the Microstructures of Cognition, D. Rumelhart, J. McClelland et al. (eds.), vol. 1, 194-281. Cambridge, MA: MIT Press. Wellek, R., and Warren, A. (1956). Theory of Literature. New York: Harcourt, Brace, and World. Wilden, A. (1987). The Rules Are No Game. London: RKP.

System and Text: Making Links

M.A.K. Halliday Sydney University Ruqaiya Hasan

40

Macquarie University

It is always refreshing to read Robert de Beaugrande: not just because he writes breezily (though that helps) but because he clears the air of stale and suffocating ideas. The picture he presents here of twentieth century linguistics, and its twenty-first century prospects, is one we can largely endorse. The prevailing feature of linguistics as a discipline has always been its shortage of data; linguists have never been able to establish the real nature of a linguistic fact to theorize the relationship between instance and system, between events that can be observed and the principles that lie behind these events. Granted that this relationship is considerably more complex with semiotic processes than it is with physical ones (Halliday 1996), it still seems as if linguistics has all this time been at about that point in its development where physics was in the sixteenth century, before technology made it possible to carry out systematic experiments. The arcane theoretical constructions that we have been familiar with in linguistics the separation of language and speech, the determinate conception of grammaticality, the idealised sentences, all de Beaugrandes pseudos and semis can be seen as attempts to make up for this lack of empirical foundation, by inventing data in the way that physical scientists had sometimes had to do before they were able, or else allowed, to observe. What has transformed the scene, or at least is now transforming it, is the corpus: an expanding repository of authentic linguistic data, spoken as well as written, which is accessible for systematic observation and research. As is so often the case, it was a particular advance in technology that made this transformation possible: in this case, the development of the computer. It is about twenty years now since computers evolved to the stage where they became available as tools for serious linguistic research; in that time the corpus has emerged as the primary resource for extending our knowledge of lexicogrammar and semantics, especially in conjunction with complementary achievements in natural language processing. We now at last have data to balance, and constrain, the proliferation of theories that de Beaugrande alludes to. Where we have to demur, however, is in coupling together Saussure and Chomsky in the way that de Beaugrande seems to be doing (cf. paragraphs 3, 10, 13, 23). Saussures reason for excluding a science of parole is that such an enterprise would be unfeasible because parole is in a constant state of flux, whereas Chomsky excludes performance because he sees it as defective and therefore of no possible interest. Where Saussure is multivocal and always leaves room for doubt (Thibault 1997 passim, Hasan 1999), Chomsky is self-assured: he deals in univocal certainties. There is no place in Chomskys scheme of things for the dialectic of langue and parole that Saussure allows for in considering language evolution and change. Thus while Chomsky would not see the absence of a body of authentic data as imposing any kind of limitation on the scope and power of linguistics, Saussure by contrast would 41

almost certainly have recognized the vast potential that would be opened up once linguists had the ability to observe their object of study, captured as it is in motion by the scale of the corpus, something that practitioners in the natural sciences had long taken for granted. De Beaugrande gives a lively account of the early Chomskyan agenda and its baleful relegation of a language to a set of (by definition grammatical) sentences; and he traces the emergence of a text linguistics as an alternative mode of understanding. The conception of a text linguistics is itself a hangover from that time, when the dogma of sentence syntax was breaking down; text was coming on to the agenda but it proved impossible to theorize text taking sentence syntax as the point of departure, whether by treating a text as a string of sentences, or by writing a special text syntax. De Beaugrande deplores this segregation of text from language; yet he seems at times to accept it: he retains the term text linguistics as in his title, and also the terminological contrast of text with language (e.g. in 21). If you oppose language to text in this way, it suggests that text is somehow a distinct phenomenon from language; we would prefer to contrast text with system, treating language as a unitary phenomenon encompassing both. What is most problematic, perhaps, is de Beaugrandes implied equation of text with language in use. This is problematic on three counts. One is the question whether all language in use is going to qualify as text; consider as an extreme example the language use of certain categories of aphasics. Secondly, and more seriously, de Beaugrande seems to leave out of consideration the properties which make a text a text. For some time now it has been generally agreed that a text is characterised by texture and structure, whatever the terms by which these properties may be referred to. Such concepts no doubt need to be further refined and elaborated; but it seems desirable not to abandon the notion that there are certain characteristics that are criterial to recognizing any given instance of language in use as a text. It might be that de Beaugrande is talking about text in the general sense of any instance of spoken or written language, rather than about 1a text as a linguistic unit. But he does refer to a text, suggesting that it is a unit which links language to speech ( 23); and that it functions in this way because, while it is an actual system, it retains some of the virtuality that is characteristic of language. We may interpret this in our terms as saying that a text is an environment within which an instantial system may emerge and grow; we would agree, noting however that this can only happen because of the systemic property of language itself. This is why instantiantion in the form of text brings about change in language a point which appears as a sub-motif in de Beaugrandes own paper. But this property of texts makes it all the more important to clarify the nature of textness in relation to the corpus. Certainly, a corpus is (or can be) a repository of texts. But this does not mean that every chunk of discourse we extract from a corpus ipso facto constitutes a text, or (and this is our third point) that attending to particular features of corpus-derived instances is the same thing as text analysis. Analysing a 42

text is not the same thing as analysing instances of language in use. We need a richer, more constrained view of what constitutes a text especially if we want to relate texts (as linguistic entities) to other semiotic modalities, as de Beaugrande is keen to do. We are now technologically in a good position to explore some of these modalities, and to show how the various modalities co-operate in creating structures of meaning. This type of semiological enquiry can enrich our idea of performance and throw light on how language is a resource for other modes of semiosis while it itself is also resourced by them. This is yet another respect in which the computer makes possible a qualitative transformation in our understanding of linguistic processes. In celebrating the computer, though, we should not forget its precursor: the thing that first opened our eyes (and ears) to the realities of language, namely the tape recorder. It was the tape recorder that began the transformation, because it opened up the world of spoken language, especially informal, spontaneous, conversational speech. This had two quite separate consequences. One was to reveal what writing left out, the rich variety of meaning potential that is built up by intonation and rhythm. We could now observe these resources at work, and investigate how they functioned, in integration with the rest of the grammar, to create meaning in text, particularly the information flow and the negotiation of interpersonal space. The other effect of the tape recorder was to reveal the intricacy and highly structured nature of the spoken language. Back in 1964 Halliday, McIntosh & Strevens (1964: 284) pointed up the false notion that speech is less grammatical than writing; the myth that spoken language was nothing more than an untidy procession of featureless fragments, incapable of analysis could no longer be seriously upheld. But the tape recorder showed up more than that: it showed the enormous semogenic potential of informal speech that here were the semantic frontiers of language, where new meanings were constantly being created and semantic systems extended and enriched. In other words, it showed us the order of speech, its choreographic complexity as each moment becomes the point of departure for a further discursive move. It might be helpful to present an example of what we mean by this characterisation. Consider the following conversation which took place at an exhibition of paintings (Svartvik & Quirk 1980: 215-8): A: Yes; thats very good. ... I wouldnt be able to have that one for some reason you see: this checker board effect I recoil badly from this. I find I hadnt looked at it, and I think its probably because it probably reminds me you know of nursing Walter through his throat, when you play checker boards or something. I think its it reminds me of the ludo board that we had, and I just recoiled straight away and thought [mm] not not that one, and I didnt look inside; but thats very fine, [mm mm] isnt it? very fine, yes. B: Its very interesting to try and analyse why one likes abstract paintings, cause I like those checks; just the very fact that theyre not all at right angles means that my eyes dont go out of focus chasing the

43

lines [yes] they can actually follow the lines without sort of getting out of focus. A: Yes Ive got it now: its those exact two colours you see, together. He had he had a blue and orange crane, I remember it very well, and you know one of those things that wind up, and thats it. B: It does remind me of meccano boxes [yes well] the box that contains meccano, actually. A: Yes. Well, we had a bad do you know; we had oh we had six or eight weeks when he had a throat which was [mhm] well at the beginning it was lethal if anyone else caught it. [yeah] It was lethal to expectant mothers with small children, and I had to do barrier nursing; it was pretty horrible, and the whole corridor was full of pails of disinfectant you know [mm], and you went in, and of course with barrier nursing I didnt go in in a mask I couldnt with a child that small, and I didnt care if I caught it, but I mean it was ours emptied outside you see [mm] and you had to come out and you brought all these things on to a prepared surgical board [mm mm] and you stripped your gloves off before you touched anything [mm] and you disinfected oh it was really appalling [mm]. I dont think the doctor had expected that I would do barrier nursing you see [mm] I think she said something about she wished that everybody would take the thing seriously you know, when they were told, as I did, cause she came in and the whole corridor was lined [mm] with various forms of washing and so on, but after all I mean you cant go down and shop if you know that youre going to knock out an expectant mother. It was some violent streptococcus that hed got and he could have gone to an isolation hospital but I think she just deemed that he was too small [yes mm mm] for the experience, and then after wed had him, you know, had him for a few days at home this couldnt be done. [mhm] She made the decision for me really, which at the time I thought was very impressive, but she didnt know me very well: I think she thought I was a career woman who would be only too glad and would say oh well hes got to go into a hospital , you know, so she made the decision for me and then said its too late now to put him into an isolation hospital; I would have had to do that a few days ago which, I thought, I didnt want her to do! B: Do nurses tend to be aggressive, or does one just think that nurses are aggressive? A: Well, that was my doctor [oh], and she didnt at that time understand me very well. I think she does now. It is interesting to speculate whether there is any other technical advance in the offing, which would complement the tape recorder and the computer in giving access to the processes of language. Is it likely that magnetic resonance imaging (MRI), or some other form of brain scanning, will develop to the point where, in observing the neuronal activities associated with speech, we can not simply note that such activity is taking place, in various different regions of the brain, but measure it and match it up with particular instances of speech? Sydney Lamb (1999) has consistently represented linguistic processes in connectionist networks that could be interpreted in analogous terms.

44

But let us come back to de Beaugrandes concern with the effect of the computerized corpus. The scale of the data available in a corpus like the Bank of English makes two kinds of enquiry possible: one quantitative and the other qualitative. On the one hand, we can conduct quantitative studies of grammatical systems, like polarity and tense, which bring out (a) general frequencies for the language as a whole (i.e. putting all varieties together), and (b) frequencies for particular functional varieties, or registers. We can then test hypotheses about single systems (e.g. that positive will be more frequent than negative, by about one order of magnitude), about register variation as the resetting of systemic probabilities, and about the probability profile of grammatical systems as a whole. (see Halliday & James 1993; Halliday 1993). On the other hand, we can use the corpus data to explore and establish what de Beaugrande here refers to as the missing links : lexicogrammatical patterns, involving colligation and collocation, which lie somewhere in between the systemic patterns recognized by grammarians of the language and the instantial patterns emerging from a single text. de Beaugrande provides illustrative examples of these; see especially his section F, intertextual actualizations of theory and practice. In what sense does de Beaugrande want us to understand missing links? They are links between two modes of systemic order, the virtual systemic order of the language, and the actual systemic order of the text. We would interpret the virtual and the actual as the two poles of the instantiation cline, which we refer to as system and instance; but we would see them as being different perspectives, different standpoints of the observer, not as phenomena of two different kinds. Of course, all linguistic patterns, even the most general systems of the grammar, are manifested in text, and have to be examined in their actual mode of being; but whereas some are very readily systemicized (networked, in the terms of a systemic grammar), others such as these missing links are more intractable, and as such they are subject to diminishing returns. So, in investigating how the words theory and practice, and their various derivations, function in construing specific domains of experience, de Beaugrande naturally locates himself at the instance end, generalizing from particular texts that share certain features in common. It would not be impossible to approach these phenomena from the system end, and eventually we shall have to do just that; but at the present moment in history to do so would consume a massive amount of theoretical (and practical) energy which we cannot expect to afford. On the other hand, the collocation of theory with practice clearly is a systemic feature of the language; we could readily determine, for example, what proportion of the occurrences of theory were within some defined collocational range of instances of practice, and we could establish the colligation of these terms, together and separately, with particular process types in the transitivity systems. Francis and Hunston (2000) also locate such colligational patterns in some middle ground between language and text, or in our terms between the system and the instance. Francis and Hunston show that,

45

while their patterns are clearly revealed by the corpus data, such patterns are simply not brought into conscious awareness either by introsepection or by the inspection of occasional samples. They then make the stronger assertion that such patterns cannot be accommodated within any general theoretical grammar. We would reject this assertion; neither the fact that they have only been retrieved from corpus analysis nor the fact that they are, or involve, regularities of a lexical kind, is sufficient reason for excluding them from the reach of a grammatical theory. Note in this connection de Beaugrandes reference to Hasans work in lexis as most delicate grammar (Hasan 1985; 1987) where Hasan sets out, and illustrates, the location of lexis within a unified lexicogrammar. De Beaugrande does not suggest that his missing links cannot be incorporated within a general linguistic theory. His intertext (cf our macrotext; also text type as a set of semantically related texts seen from the instance end) is a theoretical construct; its regularities could not be captured in terms of rules for the grammaticalness or well-formedness of sentences, but they can be understood in terms of the tuning of [...] probabilities among the texts that go to make it up (cf Hallidays 1991 account of register variation as a resetting of probabilities). Then moving to a higher level of abstraction whence to theorize the relationship between text (or intertext) and language, he proposes to define a language as a general theory of human knowledge and experience evolving in a dialectical relation to texts as a set of practices for working out the theory. De Beaugrande makes it clear that these considerations are compatible with, and ultimately take off from, the theoretical ideas of Peter Hartmann and his colleagues. We also find them compatible, and agree that a language is a general theory of human knowledge or rather, that it includes such a theory (the ideational base; cf Halliday & Matthiessen 1999); but we would want to add that it is also an enactment of human social processes, and that these two metafunctions cannot exist independently. Any text must be the product of the interaction between these two, mediated by the texturing of the whole into coherent discourse. But two questions seem to arise, in relation to de Beaugrandes recommendations, and both have to do in some sense with the overall dimensionality of language. One is the question of the distance from one end to the other of the instantiation cline. We have always been struck by the impoverished conception of natural language that is implicit in sentence-based theories; Halliday once gave a paper entitled How big is a language? (summarized in Halliday 1996), in which he wondered why this question was so seldom asked, and suggested that it was because when linguists moved from words to sentences they gave up thinking about language in paradigmatic terms. He gave as one example the English verb (verbal group in terms of systemic grammar), where even a network that was far from fully extended in delicacy specified a set of over 70,000 selection expressions. This output required only a partially ordered set of less than thirty grammatical systems; when clause types are networked to any significant degree of delicacy the number soon runs into the hundreds of

46

millions. Unless one is aware of how big a language is, in paradigmatic terms, the amount of variety that is found in a corpus of texts will always seem mysterious and even threatening; it is tempting to give up at that point and invent a notion of text that exists on a different plane of reality from language. A language is not a set of texts; but nor is it a phenomenon of a different order from text. We could call it a theory of text, in the sense in which a climate is a theory of weather; but if so it is a commonsense theory, and (as de Beaugrande has often pointed out) linguists have had a hard time coming up with a metatheory that can explain the relationship between a language, as a virtual entity (the text potential), and the observable corpus of text (whether a remnant corpus inscriptionum or an apparently bottomless Bank of English) in which this potential is actualized. In passing: let us certainly let the texts represent themselves it is a way of showing that we trust them (recalling Sinclair's injunction (1992) to 'trust the text'). Of course, once any linguistic event becomes a text for the linguist, it thereby becomes a re-presentation: it is a record of a performance, even if we are attending to it in spoken form on tape, and if it is in writing it has undergone further re-presentation a writing system is a commonsense distillation of what aspects of a text can be left out when it is preserved. Moreover as de Beaugrande points out what we call text is in fact a spectrum of possible readings. J.R. Martin prefers to take the reading rather than the text as the endpoint of the instantiation cline. The other aspect of the dimensionality of a language is its thickness: its stratal pattern of organisation. What we are calling the text is a stratified entity, its strata being linked by a relationship, called metaredundancy by Lemke (1993), which also links the text to its ecosocial environment, or context of situation. What we observe as being actualized is text in context, or reading in context if that is to be preferred. Hasan (1995, 1999) shows what is involved in representing context as part of a thicker, dimensional concept of text itself. We assume that de Beaugrande would be in sympathy with this enterprise he has written about it himself in various places elsewhere; but it is perhaps important to make the theoretical status of context explicit, because if we want, as he suggests, to invest our own status as participants, in analysing or interpreting data (in whatever guise, including Firths dictum the text means what you (sc. the linguist) say it means), we cannot do this without taking as primary data the unitary phenomenon of text-in-context. Putting these two dimensions together, instantiation and stratification, we arrive at an interpretation whereby a text, in its context of situation, is an instance of the linguistic system, in its context of culture. If we are to theorize the interpretation of the social and the semiotic (as is essential if de Beaugrandes project for education is to have any hope of succeeding), then in order to bring real language back into the centre we will need to be surveying very large corpora of actualized texts (.97) in relation to their social contexts: not only the instantial (situational) contexts of the texts themselves but also the context

47

of the corpus as a social activity its relation to the differential distribution of textual practices across society and to the way the reading of a text is inalienably linked to the social positioning of the reader. De Beaugrande is outstanding for his insistence on the social accountability of the linguist. What we are emphasizing here is the importance of embodying this commitment into the theorizing of language itself: assuming a text linguistics at, and beyond, the millennium, the new freedom that comes with access to large quantities of authentic data will become a truly democratic force only to the extent that the data are matched by a socially accountable theory of semiotic systems and processes. References Francis, Gillian and Hunston, Susan.2000. Pattern Grammar: A corpus-driven apporach to the lexical grammar of English. Amsterdam : Benjamins. Halliday, M.A. K. 1991. Towards probabilistic interpretations. In Eija Ventola (Ed.) Functional and Systemic Linguistics: Approaches and Uses. Berlin : Mouton de Gruyter. Halliday, M.A.K. 1993. Quantitative studies and probabilities in grammar. In Michael Hoey (Ed.) Data, Description, Discourse: Papers on the English Language in honour of John McH Sinclair. London : Harper Collins. Halliday, M.A.K. 1996. On grammar and grammatics. In Ruqaiya Hasan, Carmel Cloran & David Butt (Eds.) Functional Descriptions: Theory in Practice. Amsterdam : Benjamins. Halliday, M.A.K., McIntosh, Angus & Strevens, Peter. 1964. The Linguistic Sciences and Language Teaching. London : Longman. Halliday, M.A.K, & James, Z L. 1993. A quantitative study of polarity and primary tense in the English finite clause. In John McH Sinclair, Michael Hoey & Gwyneth Fox (Eds.), Techniques of Description: Spoken and Written Discourse. London : Routledge. Halliday, M.A.K, & Matthiessen, Christian. 1999. Construing Experience through Meaning: A Language Based Approach to Cognition. London : Cassell. Hasan, Ruqaiya. 1985. Lending and borrowing: From grammar to lexis. In John E Clark (Ed.) The Cultivated Australian. Amsterdam : Helmut Buske. Hasan, Ruqaiya. 1987. The grammarians dream: lexis as most delicate grammar. In Robin P Fawcett & M A K Halliday (Eds.) New Developments in Systemic Linguistics, Vol 1: Theory and Description. London : Francis Pinter. Hasan, Ruqaiya. 1995. The conception of context in text. In Peter H Fries & Michael Gregory (Eds.) Discourse in Society: Functional Perspectives. Norwood , NJ : Ablex. Hasan, Ruqaiya. 1999a. The disempowerment game: Bourdieu and language in literacy. Linguistics and 48

Education 10(1): 25-87. Hasan, Ruqaiya. 1999b. Speaking with reference to context. In Mohsen Ghadessy (Ed.) Text and Context in Functional Linguistics. Amsterdam : Benjamins. Lamb, Sydney. 1999. The Pathways of the Brain: The Neurocognitive Basis of Language. Amsterdam : Benjamins. Lemke, Jay. 1993. Discourse, dynamics and social change. Language as Cultural Dynamic, Special Issue of Cultural Dynamics 6.1-2: 243-275. Sinclair, John McH. 1992. Trust the text: the implications are daunting. In Martin Davies & Louise Ravelli (Eds.) Advances in Systemic Linguistics. London : Pinter. Svartvik, Jan & Quirk, Randolph (Eds.). 1980. A Corpus of English Conversation. Lund : C W K Gleerup. Thibault, Paul J. 1997. Re-reading Saussure: The Dynamics of Signs in Social Life. London : Routledge.

Grammar versus Text Wolfgang U. Dressler University of Vienna

Since texts are among the most complex linguistic signs (or even units) which are objects of linguistic research, the number of promising approaches to their investigation is much higher than with less complex signs (e.g. word or sentence). As a consequence, even Beaugrande's very rich contribution cannot do justice to the wealth of competing promising approaches to text, be it in the ending or (even more) in the upcoming millennium. In accordance with the available space, I will focus my response just on one avenue towards studying the interconnected relations between grammar and text, and between virtual and actual systems, and in doing so I must refer to some of my previous work for further information. I quite agree with Beaugrandes vision of text (section 23) as a unit which actually links language to speech, but I disagree with his equation or near-equation of language and speech with Noam Chomsky's competence vs. performance or Ferdinand de Saussure's langue vs. parole. Already Louis Hjelmslev and Eugenio Coseriu have argued that this dichotomy must been enlarged to trichotomy which includes language as a social institution = linguistic norms as mediating between language as a virtual system and actualisation in performance/parole.

49

Hjelmslev nowhere says so; on the contrary, he vowed that if social conditions become the content of conventional linguistics, we will fail to grasp the totality of language, and incur the danger of overlooking language itself (1969 [1943]: 4f ). That this trichotomy works well for those systems of grammar, also within my own naturalness approach (cf. Dressler 1999a), I have shown in many places, e.g., for phonology in Dressler (1984), for morphology in Dressler (ed. 1987). It works well also for the lexicon, e.g., when differentiating between 1) potential meanings of a word within (usually open) system, 2) the institutionalised meanings of this word, as they should be found in a good dictionary, and 3) the actual meanings in a running text. Clearly, such a conceptualisation renders the application of the empirically very important dichotomy between type and token more difficult than practitioners would like, i.e. much (explicit or implicit) theoretical groundwork is needed for operationalising the type-token relation. This is most obvious on the text level, e.g., if we want to compare tokens of the type love letter (or obituary, etc.) and count the occurrences of definitional or of proto typical properties as against accidental properties of such a type. These types, which represent genres (German Textsorten) are clearly to be assigned to the level of language as institution (norm) Now can we construct for these genres also correspondences on a level of the virtual system of language? Of course we can talk about virtual love letter (or obituary) and its distinctive Saussurian value in opposition to other genres of the same language. But clearly such virtual concepts are much vaguer and more fluid than any phonological, morphological, syntactic, and lexical unit of the virtual system of the same language. This is the immediate main reason why attempts at characterizing non-texts and constructing text grammars have been futile (see Beaugrande 18, 37). Thus too,stars (asterisks, a tool introduced by generative grammar) are much dimmer in text than in grammar. As further consequences, there have been modular approaches to grammar, whereas nobody has ventured to postulate a specific text module. Accordingly, the complexity of a (virtual) grammatical system can be described in terms of a Building-Block Model or of a part-whole model of complexity theory, whereas, as shown by Beaugrande himself, the complexity of text has to be accounted for in quite different terms. Therefore it has been a scientific progress, even a must (pace Beaugrande) that grammar, which had been firmly connected with logics and rhetorics in the classical trivium since medieval times, has been isolated from text and posited as a separate object of study (cf. Dressler 1999b). This view, however, entails that grammar cannot be the core of linguistic phenomena with texts as superficial epiphenomena; rather, text linguistics, discourse analysis and pragmatics deal with the essential language phenomena, whereas virtual linguistics systems of which virtual grammar is the most systematic and structured part represent secondary idealization. These idealizations should not be seen only as models. Rather they correspond in a way which linguists interested in the psychological reality of their models must elaborate and refine, to what speakers

50

construct themselves. Similar to constructivist frameworks which describe the self-organizing emergence of complexity in physical, chemical and biological systems, constructivist and emergentist models of language acquisition must describe how children construct and reconstruct, step by step, correspondences to the complexity of their target/input language. This happens first on the levels of phonetics on the one hand, of text and pragmatics on the other (cf. Dressler 1996). Later on children also construct fragments of grammar (Dressler ed. 1997). All these constructions are patterns which appertain to the level of norms which of course differ from institutionalised adult norms. But when children become aware of the productivity of grammatical rules, then they (re)construct a further level of reader mechanisms of a virtual system, which represent the core of virtual grammar. Properly textual phenomena lack such a core and thus, what children, adolescents and adults can construct in this area, is a rather loose system of potentialities. For this system Beaugrande (30) is correct when concluding that virtuality is by definition open and dynamic. As a result of these views, I cannot agree with Beaugrande's (17, 31) characterizations of grammar as frozen (which recalls Leo Spitzers famous identification of grammar with frozen style) nor with his rejection (32) of real synchronic system: of course, performance is always dynamic (even in the severest syndromes of global aphasia), and the level of language as institution includes many norms which are elastic and full of social transitions. Beaugrande is also correct when thinking of the open virtual systems of text. But the constructions of virtual grammar are much steadier. This view on the difference between text and grammar (especially qua virtual systems) engenders differential predictions for language impairments, e.g. in aphasia (cf. Dressier & Stark eds. 1988). It correctly predicts impairments of both performance and norms of both grammar and text, which is to say no difference. Aphasics may even construct new textual norms during therapy, as Dressler et al. (1990) have shown for gender-specific discourse. But whereas there is some evidence for the impairment of entities or properties of virtual grammar, there is none so far for impairment of virtual text. Quite predictably, the steadiest part of virtual text systems occurs at the interface of text and grammar, where rules of e.g. co-reference are constructed, whose status comes close to, but is not identical with, that, of syntactic rules of sentence co-reference. Not surprisingly, this subfield of text linguistics has been called text syntax or transphrastic syntax in the s1970s. Text linguistics in the coming millennium would do well to distinguish this (rather marginal) subfield from the main objects of textual investigations. References

Dressler, Wolfgang U. 1984. Explaining Natural Phonology. Phonology Yearbook 1. 29-52.

51

Dressler, Wolfgang U. 1987. Leitmotifs in Natural Phonology. Vienna : sterreichischen Akademie der Wissenschaften. Dressler, Wolfgang U. 1996. A note on rhematic disagreements in early child language. In B. Partee & P. Sgall eds. Discourse and Meaning. Amsterdam : Benjamins, 205-219. Dressler, Wolfgang U. 1999a. On a Semiotic Theory of Preferences in Language. The Peirce Seminar Papers 4, 389-415. Dressler, Wolfgang U. 1999b. Textlinguistik und die Isolierung der Grammatik von Pragmatik und Diskurs. In F. Fuerbeth et al. eds. Zur Geschichte und Problematik der Nationalphilologien in Europa. Tbingen: Niemeyer. 823-832. Dressler, Wolfgang U. & Jacqueline A. Stark. 1988. eds. Linguistic Analyses of Aphasic Language. New York : Springer. Dressler, Wolfgang U., Ruth Wodak and Csaba Pleh. 1990. Gender-specific discourse in aphasia. In Y. Joanette & H. Brownell eds. Discourse Ability and Brain Damage. New York : Springer, 236-245.

Response to Robert de Beaugrande

Roland Harweg Ruhr-Universitt Bochum

The central object of a linguistic discipline which calls itself text linguistics is the text. It is, therefore, to be expected that in an article on text linguistics the notion of text should form the main topic, and in a certain way this expectation, in Beaugrandes article, is clearly fulfilled. Nevertheless, I must confess that I have no clear idea of what Beaugrande means by this notion exactly. What I presume is, however, that his notion and idea of text is at least in part different from the one I have. Basically speaking, one can, it seems to me, distinguish between two different notions of text -- two notions of text which in the view of many linguists (and this is what makes the matter rather complicated) overlap or even coincide. They coincide, however, only from a certain point of view, namely extensionally, not intensionally. Intensionally, they can clearly be distinguished, at least by opposing them to their opposite notions. Thus, the opposite notion of one of the two notions of text is the notion of language, whereas the opposite notion or, more precisely, the main opposite notion of the other of the two notions of text (for this one has more than only one opposite notion) is the notion of sentence.

52

The former of the two notions of text designates a certain form of manifestation of language in a wider sense and is as such roughly synonymous with notions like parole and performance, its opposite notion being language in the narrower sense, i.e. langue or competence, respectively. The latter of the two notions of text, on the other hand, designates a certain level (in my conception the highest) in the hierarchy of linguistic units, a level within that hierarchy which consists, among others, of the levels of the morpheme, the word, the sentence, and the text. But although the two notions of text are, intensionally, well-distinguished, in the eyes of many linguists they, as was already said, fall together from an extensional point of view. Thus, for Chomsky and most of his adherents, the sentence, as is known, is the highest hierarchical level of language which still belongs to language in the narrower sense, i.e., to competence, and for Saussure and many of his followers, exceptions set aside, already the word is the highest linguistic unit which still belongs to language in the narrower sense, i.e. to langue. Texts, in the view of Chomsky and most generativists, belong to performance, and in the view of Saussure and his followers they belong to parole. What is the intensional opposite notion of the notion of text in Beaugrandes article, is it language in the narrower sense, i.e. the langue or is it the sentence? Maybe a wholly unequivocal answer to this question is not possible, but there is evidence for the assumption that this opposite notion be language (in the narrower sense), i.e. the langue. At least, this opposite notion is for him the main, the dominant opposite notion of the notion of text. Thus, the first chapter of his article bears the heading Language and text in modern linguistics, and though its terms language and text do not occur in the headings of the following chapters, the headings of the first two, namely Virtual system and actual system and Theory and practice, obviously point in the same direction. The first and the second sentence of these chapters, respectively, are even explicitly clear about this. Thus, Beaugrande in the first sentence of the former of these two chapters writes: As a provisional strategy, some text linguists (myself included) have for a number of years been proposing to view the relation between a language and a text as one between a virtual system and an actual system; and in the second sentence of the latter of these two chapters he writes: we could define a language to be a general theory of human knowledge and experience evolving in a dialectical relation to texts as a set of practices for working out the theory. Of these two characterisations of the opposition between language and text Beaugrande considers the former as merely provisional and the latter as situated on a higher level. Nevertheless, he also retains the former one. This is shown by his use of virtualistic terms like collocability and colligability on the one hand and actualistic terms like collocation and colligation on the other, terms he uses in order to describe basic phenomena of language on the one hand and basic phenomena of texts on the other. His higher evaluation of his characterisation of the two phenomena by means of the opposition

53

theory vs. practice is based on the mutual dialectical connection which consists between theory and practice. The opposition to language (in the narrower sense) is, it is true, the main opposition in which Beaugrande sees the phenomenon of the text, but not the only one. He sees the text, as some of his remarks betray, partly also in opposition to the sentence. However, he sees this opposition, whenever he speaks about it, connected with another opposition, namely that between actuality or authenticity on the one hand and inventedness on the other. Thus, he speaks, whenever he speaks of sentences or more precisely, of isolated sentences, of invented sentences, and when he speaks of texts, he speaks of actual or authe ntic texts, and the adjectives he uses in this are, in the last analysis, only other expressions for the opposition between langue and parole, between virtualness and actualness. For Beaugrande, the opposition between language and text (in the sense of langue and parole, respectively) and the opposition between sentence and text are not independent from one another. True, he admits that, for instance in speech act theory, one has attempted to stretch and bend the concept of the sentence in the direction of performance, and that Hjelmslev, for instance, has tried to integrate the text into the langue; but, he asserts, Hjelmslev at least has not been able to show how this could be carried out (cf. 8). The opposition between sentence and text, though perhaps in part as a result of his discussion of the history of modern linguistics, in fact does occur with Beaugrande, but it does so only inextricably interwoven with the opposition between language and text and increasingly eclipsed by this opposition. Finally, it is even wholly abandoned. Consider, in this connection, two passages in Beaugrandes article. In the first of these two passages, Beaugrande still unequivocally proceeds from the conception that the text is a sequence of sentences ( 14), a conception which still abides by the opposition between sentence and text, but he turns this conception the other way round, such that he prefers to say that the sentence is a segment within the text ( 15). The difference is, as far as the mere relationship between sentence and text is concerned, merely one of perspective, but for Beaugrande this difference of perspective implies much more. He considers the sentence as a unit of the langue and that he fears lest, when he proceeds from the sentence to the text, that the treatment of the text would ascribe the languespecific properties of the sentence to the text as well. Inversely, his proposal to consider the sentence as a segment of the text, i.e. to proceed from the text to the sentence, could confer the parole-specific or performance-specific treatment of the text (which he presupposes) to the sentence. Even there where he believes in a clear-cut hierarchical opposition between sentence and text is not ready to conceive of this opposition without the opposition between language and text (in the sense of langue and parole or competence and performance) and, what is more, that he subordinates the former to the latter. The opposition between sentence and text, which in this passage appears solely as being subordinated to

54

another opposition, in a following passage ( 59) appears to be downright abandoned, and the abandoned opposition itself is, as it were retrospectively, even watered down. It is watered down by the fact that the conception of a text as a sequence of sentences is supplemented or even replaced by the conception of the text as a mere set of sentences. True, in principle the notion of set might be in fact neutral as over against -- in Hjelmslevian terms -- paradigmaticity and syntagmaticity, but as an alternative to the notion of sequence it is merely to be understood paradigmatically. In this conception, however, the conception of a text as a set of sentences does not make allowance for the traditional hierarchical opposition between sentence and text. But this conception is, just as the conception that the text is a sequence of sentences, a conception which Beaugrande wants to be replaced by another one. This other conception is the conception of the text as a contribution to intertext ( 59). What, however, is the or one intertext? So far as I can see, the notion of intertext is not explicitly defined by Beaugrande. In one place, in 35, he seems to identify the notion of intertext with that of conversation, and although he refuses the conception of an indefinitely long intertext and of its practical correlate, possibly a conversation in which the last word can never be said ( 35), he at least exemplifies his concept of finite intertexts by the phenomenon of conversations -a conception that reminds me of Walter A. Kochs (1971: 260ff.) concept of conversations as, as he has called them, bi-textemes or n-textemes. In this conception, the intertext appears to be a unit which, in the syntagmatic dimension, is superordinate to the text; for in this interpretation the relationship of the intertext to the text seems to be similar to that of the text to the sentence. There are, however, still other contexts in Beaugrandes article, contexts in which the intertext is not conceived of as a higher syntagmatic unit, but as a higher paradigmatic unit, namely as a set of texts -- so, for instance, in 89, where, in connection with the notions of intertext and intertextuality, Beaugrande speaks of large numbers of texts, but already in 35 as well, where, in the immediate context of the identification of intertext and conversation, Beaugrande speaks of the discourse of history. Just as with the rejected definition of text as a sequence or set of sentences, so also in the case of the concept of the intertext: the opposition between syntagmaticity and paradigmaticity seems to play no role for Beaugrande. Apart from this, the concept of paradigmaticity, at least implicitly, seems to play the greater role, so above all when Beaugrande, as in 59 and 89, uses this concept in the singular -- a usage which, especially in connection with the notion of networks, reminds of the notion of internet and its globality and which seems to correspond to my notion of the universe of texts (Textkosmos) (cf. Harweg 1979 [1968]: 143ff.). For me, in contradistinction, the opposition between syntagmaticity and paradigmaticity plays a great role. Thus, I distinguish between the universe of texts (Textkosmos) as the paradigmatically organised set of all texts and, inside this (unilingual, plurilingual or omnilingual) universe of texts, the different

55

classes or sorts of texts on the one hand and the syntagmatically organised macrotexts (Grossraumtexte) (cf. Harweg 1970). These macrotexts, however, are not identical with Beaugrandes conversations or Kochs bi-textemes and n-textemes. These latter texts are, in my conception, still normal texts (Kleinraumtexte), since in my conception a plurality of speakers does not automatically produce a unit which transcends the unit of the text. For me, macrotexts are rather sequences of texts which, though separated from each other by lesser or greater temporal intervals, are nevertheless linked to each other anaphorically, i.e. in such a way that subsequent texts within a macrotext presuppose preceding ones in this macrotext. In special cases such a macrotext can also be a dialogue. This, for instance, is the casewhen the dialogue consists of an exchange of letters (between two persons) in which subsequent letters refer anaphorically to preceding ones. Another example of macrotexts are successive (consecutive) newspaper reports on the successive phases of an event developing and lasting over a certain stretch of time. The small role which is played by the Hjelmslevian opposition between syntagmaticity and paradigmaticity on the levels of text and intertext as well as the implicit preponderance of the paradigmatic on these levels manifests itself also in Beaugrandes practical methods. In attempting to exemplify his method he, in accordance with his theoretical principles, has ample recourse to concrete and authentical texts -- a procedure which is to be highly praised and whose absence in Firth and, above all, in Hjelmslev Beaugrande justly deplores. The texts, Beaugrande ( 60) says, stem from the Bank of Englis at Birmingham University, the worlds largest computerised data corpus (...) containing over 329 million words of running text. But what are 329 million words of running text? Surely they do not constitute one single running text, but just a host of running texts. Yet how many? And how many of these running texts are mere excerpts of authentical texts and how long are these excerpts? And even if the texts would not be excerpts, Beaugrande would have had to epitomise them at the latest when coming to use them for his exemplificatory purposes -- mainly purposes concerning problems of collocation and colligation, of collocability and colligability in the framework of sentences -- and actually this is what Beaugrande or more precisely, Mr Bullon, who provided him with samplings of roughly 200 characters ( 91), has done. As the collocations investigated by Beaugrande, the syntagms of the type If I were you, theory and practice, and theoretical and practical, show, the authors interest in syntagmaticity, in this article, seems to be reduced to syntagms of less-than-sentence length. To be sure, the samples are a little bit longer, but only a little bit. This seems to reveal that Beaugrandes interest in and concern for textlinguistics is not -- or at most to a very small degree - guided by a concern for the dimension of syntagmaticity, but by quite another principle, namely that of authenticity. What he has exemplified in this article, reminds me strongly of what, for a long time, many grammarians and lexicographers have done, especially (and by necessity) in the description of languages like Greek and Latin, i.e. of dead languages. To be sure, many of the descriptive goals envisaged by Beaugrande 56

may be much more delicate and, in the long run, also much more comprehensive than theirs, but they nevertheless still belong to sentence grammar and lexicography. They do not, at least if we judge them from the kind of examples discussed by Beaugrande in the article under discussion, refer to problems of coherence or cohesion of larger parts of texts, not to mention larger texts as wholes. Beaugrandes main interest in textlinguistics, to say it once more, seems to be the relationship between language in the narrower sense of the word, namely langue or competence, and text in the sense of parole or performance, and as to the very essence of this relationship he really has to offer some new and important proposals. At first sight, judging from the emphasis he lays on the authenticity of texts, one might be led to conclude that, in a certain way, he would only repeat what former linguists of the twentieth century used to claim, namely that language is a virtual phenomenon, a phenomenon of competence, and the text an actual phenomenon, a phenomenon of performance. However, what Beaugrande really does is to attack or at least to modify this wide-spread view, pointing -- justly to my mind -- to what he calls a bizarre implication of this view, namely the implication that using a language in speech triggers an abrupt catastrophic transition from stable and integrative order over to unstable and disintegrative disorder ( 22). This does not mean that Beaugrande would be ready to give up the distinction between language and text or langue and parole altogether. What it means is rather that he sees language and text nearer to each other. Thus, what he favours is a text-like view of language, and a language-like view of texts ( 55). Important for textlinguistics seems to me above all his view that texts are not wholly deprived of systemic or virtual aspects ( 23) and that texts are not, as is normally believed and sustained, exclusively created by an actualisation of lower-level virtualities ( 23). Nevertheless, this is easier to assert than to prove, and I doubt that Beaugrande, in his article, has given convincing evidence to this claim. Thus, the samples of sections or portions of authentic texts he discusses centre, as I have already said, around phenomena which, to my mind, do not reach the level of text in the sense of hierarchical order, and as to his claim that some powerful evidence for the virtuality of texts would be that a single text can be received and interpreted (actualised) in more than one way by different participants in a text-event or even by the same participant at different times ( 24), I must confess that I cannot find it convincing either. True, this possibility is some kind of virtuality, but I doubt that it is the same kind of virtuality which we normally connect with language. Moreover, the impact of this possibility, at least for normal texts, seems to be smaller than is often believed. Not by chance, Beaugrande in this connection points to literary texts ( 24). But to compare literary texts on that account, as already Wellek and Warren (1956) have done, to langue and only their individual realisations to parole seems to me somewhat hasty. This is not to say that I for myself am wholly against such an interpretation, but I think this assignment needs further specification. In principle, I must say that I heartily welcome the view that texts might be assigned both to langue and to parole, and what concerns literary texts I, too, regard them as particularly suited for this double 57

assignment. But I do this on the basis of a slightly different conception of langue and parole. Parole, in my interpetation, is defined by communicativity, langue instead by non-communicativity -- a view which implies that language is not communicative in its totality. Language is communicative only when it is actualised, and there are even actualisations of language -- I call them performance langue -- which are not communicative, so, for instance, if somebody utters some sentences only in order to test his husky voice. Now, as to literary or more precisely, fictional texts: they seem to me to belong to langue as long as they are viewed as productions of their non-fictive authors, that is, as real fiction, whereas they seem to me to belong to parole -- and this applies in particular to narrative fictional texts -, as soon as they are read as productions of their fictive narrators (who, together with the texts, are created by the non-fictive authors); for only when looked upon as produced by these fictive narrators and as received by recipients who are fictive as well or who have fictivised themselves they are communicative; when they are viewed as produced by their non-fictive writers they are, apart from interpretations on some higher, e.g. symbolical level, non-communicative. Some fictional texts, namely certain lyrical poems, even seem to be non-communicative in every regard, for they are, in the original Greek sense of the word poem, interpretable only as productions of their poets, of poets who have no fictive counterpart in the sense of fictive narrators (cf. Harweg 1992). Of course, text linguists should not content themselves with seeing a possible languicity of texts in their mere non-communicativity -- a non-communicativity which, by the way, is also to be found in the field of non-fictional texts, especially in letter-writers guides and colloquial language guides (for these are communicative only when they are used for communicative purposes) --, text linguists should also try to discover langue-specific features of texts that may reclaim the status of textgrammatical rules or which at least may be interpreted as rule-like to a certain extent. These could be roughly features of the kind Beaugrande discovered in investigating and interpreting the samples he has put forth in his article, except for the fact that these features, though stemming from real texts, nevertheless fall within the scope of sentences. Rules or rule-like phenomena which go beyond the sentence are, for instance, intersentential anaphorical relationships in the widest sense or the still unknown features thar contribute to the thematical unity of texts and their internal hierarchisation. Methodologically, Beaugrandes greatest concern seems to me his claim that the text linguist should proceed from authentic material instead of inventing his material. Although I think this might not be necessary on the level of sentences I regard his claim as sound and justified on the level of texts; for to invent texts as objects of linguistic demonstration, and that is: to write texts that have no communicational aim and value, but sound natural and at the same time satisfy specific requirements of linguistic demonstration seems to be a job that, in the long run, might be too difficult to realise. So, for obtaining reliable insights having recourse to authentical texts might indeed be indispensible. On the other hand, however, I think that the value and necessity of this method is restricted to the merely 58

heuristic phase of this analysis procedure and should end with it. For after this the exploitation of our competence should come into play. Not to make use of this competence would be a waste of reliable and valuable resources. Beaugrande, at least in principle, seems to share this opinion. True, he doubts that the native speakers intuition might be reliable for telling you what they would say until there arises a real-life occasion when they do say it ( 31) (which, for shorter sentences, I doubt), but with regard to the post-festum-use of our linguistic intuition he seems to hold the same view as I do; for he writes: Intuition operates most smoothly after the fact in making sense of what has already been said ( 31). As to the extent to which we should make use of our intuition after the fact, however, we might differ. Whereas Beaugrande seems to be willing to make use of our intuition only for making sense of what has been said, I should like to make use of our intuition also for deciding whether what has been said is acceptable, i.e. agrees with our intuition. Beaugrande, not very confident about our intuition, seems to think it necessary to consult a vast number of authentic texts in order to find out which among various variants is the more or most probable one, whereas I, more confident about our intuition, do not think that we need such a vast number of authentic texts; for my concern in authentic texts is a heuristic, not a probabilistic one. Instead of comparing a vast number of short samples, I should find it more interesting and rewarding to analyse fewer examples of greater or even great length, including macrotexts with temporal intervals between their parts. On the whole, Beaugrandes conception of text linguistics at the millennium border is a very bold and far-reaching one and seems to transcend the limits of linguistics proper (including text linguistics) in various directions. Beaugrande goes so far as to define a language as a general theory of human knowledge and experience evolving in a dialectical relation to texts as a set of practices for working out the theory ( 44). What regards language -- in contradistinction to texts -- this definition reminds me somehow of Humboldts (1830-1835) and Weisgerbers (1953-1954) views of a language as a specific view of the world, a view articulating and organising the worlds original conceptual chaos and thus transferring the world into the property of the mind. But unlike Humboldt and Weisgerber, Beaugrande adds to language the realm of texts and sees, as he says, language in a dialectical relation to texts as a set of practices for working out the theory ( 44) language is considered to be. In accordance with this view, Beaugrande tends to interpret the textual samples he analyses not only from a linguistic point of view, but also from the view-point of various other fields, such as education, sociology, and politics. The comprehensiveness of his view might have been triggered by the breath-taking vista opened by the magic date of the beginning of the new millennium already evoked in the very title of Beaugrandes article. To my mind, this comprehensiveness, though surely impressive, is somewhat too vast and runs the risk of melting the gist of text linguistics in interdisciplinarity.

59

References Harweg, R. (1979 [1968]). Pronomina und Textkonstitution. Munich : Fink. Harweg, R. (1970). Zur Textologie des Vornamens: Perspektiven einer Grossraumtextologie. Linguistics 61: 12-28. Harweg, R. (1992). Communicative and Non-Communicative Language. In: Signs of Humanity / Lhomme et ses signes, M. Balat and J. Deledalle-Rhodes (eds.), vol. 1, 187-194. Berlin : Mouton & Gruyter. Humboldt, W. (1830-1835). ber die Verschiedenheit des menschlichen Sprachbaues und ihren Einfluss auf die geistige Entwicklung des M enschengeschlechts. Koch, W. A. (1971). Taxologie des Englischen. Munich : Fink. Weisgerber, L. (1953-1954). Vom Weltbild der deutschen Sprache. Dsseldorf: Schwann. Wellek,R. and Warren, A. (1956). Theory of Literature. New York : Harcourt, Brace, and World.

Intertextuality and the Project of Text Linguistics Jay Lemke City University of New York

In his stimulating contribution on Text Linguistics at the Millennium, Robert de Beaugrande (this issue) urges us to seek out the missing links between language as a system of possibilities and text as a unique instance of the social act of speaking or writing. He seeks a rededication of the project of modern linguistics to understanding the meaning of texts in terms of a language system used for social ends in social contexts. And he re-establishes our optimism for such a project in the potential of large, computeraccessible corpora of text data to inform our investigations of what and how texts mean. My own studies of text-in-context over many years (e.g. Lemke 1983, 1985, 1990, 1991, 1994, in press) lead me to broad agreement with de Beaugrande's aims, and with the theoretical view of the textlanguage relationship that he presents. Coming to similar conclusions by a different path, my angle of view may help some readers of TEXT to grasp more of the multi-dimensional whole that de Beaugrande's conception of text linguistics seeks to define. I wish here to highlight several important points , de Beaugrande has made, and amplify them with brief arguments and examples from my own work. At various points I will raise some of the questions within this broad view of text linguistics that continue to perplex me. 60

Perhaps no question in text linguistics is more important, or more vexed, than how best to conceptualize the relationship between system potential and textual instance, whether as langue vs. parole, competence vs. performance,-- or virtuality vs. actuality. Like most of us who seek to understand how texts mean, I have had to grapple with this issue again and again in the evolution of my own work (particularly Lemke 1985, 1991, 1994, in press). By and large I have followed the general view of Michael Halliday (1977, 1978), adopted in part by Beaugrande as well, that system potential -- language as patterns of formal order inferred from some of the many regularities construable across texts -- tells us what can or perhaps cannot be said or written in a community while the instantial text tells us what has been said or written on some unique situated occasion of human social activity. This is a bare beginning. As de Beaugrande notes, texts do not entirely give up their power for meaning-making, they are not entirely meanings made. There is still something language-like, a resource potential for making further meaning, using features of a given text or of many similar texts as a tool to create new meanings in new texts. Texts are in some sense made from other texts (intertextuality) as well as from language as code. Likewise, the meaning-making potential of language as a cultural system in its broadest sense is not exhaustively described either by the purely paradigmatic systems of Halliday's lexicogrammar, nor by the purely syntagmatic syntax of the MIT school. Language as systematic resource for the making of textual meanings retains the key feature of what it makes: an inner dialectic between choice and sequence. The choices are not the same once the sequence has progressed to some particular point in the making of a text; the choices made up to now influence the possible sequences that can follow, and -- do so on multiple scales. Neither language nor text is ever a truly synchronic phenomenon; both are dynamic in nature. Whether we imagine the act of writing or speaking as an on-going production, contingent at each next moment, or the effort of interpretation; similarly, the meaning of text is, made through time, and never wholly predictably: the aggregate probabilities for each choice that are the system are not only re-weighted for each situation and each text, but dynamically shift during the process of text-production itself. The overall meaning potential of the language, whether as grammar alone or as the totality of its semantic resources, is also constantly changing, on many timescales. I consider myself, when necessary, a semiotic materialist; that is, I take the basic reality of language to be the sounds and marks by which we invoke its power to mean, but their semiotic power I take to reside not in the sounds and marks themselves, but in the particular conventional systems of meaning relationships among them that are repeatedly construed by the members of some community. Thus the material reality of language for me is in the texts themselves as material objects (or material dynamical processes), and their full semiotic force appears when they are interpreted into unique, wholly definite meanings, however unstably. What we call grammar or syntax, semantic systems, language rules, etc. are all the products of our analysis, mere theoretical conveniences, whose only real existence is in the texts we ourselves produce as linguists and in which they are embedded. I am perhaps at the philosophical 61

margins in believing that no such relationship we construe is an absolutely necessary one; we could imagine and speak of the world, and survive, in indefinitely many ways. Our being-in-the-world constrains something about how we talk and behave survivably, but it does not determine any particular meaning relationship absolutely. I say this because such issues do matter to the classic top-down vs. bottom-up debate over the text-language relationship. Platonists and neo-Cartesians grant a causal reality to formal abstractions; I do not, and neither as I read him does de Beaugrande. All that we say of language is an abstraction from our experience with particular texts, and is itself just one more text. So the problem of the missing links then becomes a question of appropriate units of analysis and useful levels of abstraction as we move from the unique meaning-interpreted-text-in-context, to features of texts that are typical or frequently repeated under some conditions, to features that seem ever-present and only very slowly changing over historical time. Perhaps it is not surprising that modern scientific linguistics looked first to identify the most invariant features across all texts, and construed first phonological and then morphosyntactic patterns. Nor is it really surprising, in the model I would share with de Beaugrande, Halliday and many others, that these patterns enable us to make textual meaning, but do not very much constrain the particular meanings we make, and so they serve to tell us relatively little about what particular texts mean. I am an applied linguist; I am more interested in the texts I study, and what they tell me about the society I live in, than I am in the details of linguistic abstractions for their own sake. I turn to those details only to help me understand better the possible meanings of various texts in their social contexts. When I first studied the spoken texts made by teachers and students in science classrooms (Lemke 1990), I was struck by their affinities to and differences from the written-texts of professional science which I had studied as a physicist. I was also impressed by the very many ways in which what counts for the scientific community as the same scientific explanation can be presented in words. The critical meanings in these classrooms were matters of intertextuality: How could students synthesize from ten partial explanations one coherent understanding? How can we learn to judge which differences of wording matter to the scientific content of a text and which do not? My answer to these puzzles was a missing link', which I called an intertextual thematic formation (Lemke 1983, 1994). It abstracted from a set of thematically related texts their common semantic patterns insofar as these mattered to a particular community for a particular set of social purposes. It recognized the role of grammar and textual cohesion, but it was far more local', more register-specific. When I attempted to understand the reverse process, going from such a common textual pattern to any actual wording that conformed to it, I immediately saw how much else texts are about besides their topical content (social relationships, evaluative orientations, rhetorical organization, etc.). As I investigated these other meaning functions of texts, I began to appreciate why we make texts at all.

62

Why should we not be content with sentences of a few clauses? What kinds of meanings can be made with longer texts that cannot be made with shorter ones? How do the short-scale meanings of texts add up to their longer-scale meanings? De Beaugrande notes in his discussion of the practicalities of corpusbased analyses that in many registers and genres of text we cannot have much confidence in the specificity of the meaning of even a short phrase like theory and practice', unless we look across some minimal scale of co-text. In my own analyses of how texts present an evaluation of the warrantability, desirability, importance, expectedness, etc. of a semantic item (word, more often phrase or clause), I found that in many cases both the polarity and the degree of the evaluation had propagated grammatically and rhetorically from a distance of several clauses or even paragraphs in the text (Lemke 1998). Texts are organized on many formal and functional scales, and these matter very much to how they make their meanings. Our intermediate units of analysis, our missing links must take them into account to be useful for our purposes. it is a basic feature of text that it is semantically heterogeneous; it makes different, if related, meanings at different points. This is the visible or audible trace of the dynamical processes of its production and interpretation, processes which generate new and different next meanings from whatever meanings have already been made up to some point. As we abstract upwards from a set of texts toward common patterns, we must be guided by some social purposes of our own community. And we always are, knowingly or not. There are just too many possible patterns, especially close to the level of texts themselves; only interest or need enables us to selectively foreground some rather than others. In both his general arguments and his textual examples, de Beaugrande insistently reminds us of the politics of linguistics as an intellectual endeavor. We must indeed subject our own professional texts to the same scrutiny we direct at others'. The register of linguistics, the common genres and rhetorical commonplaces of linguistics, the covert heroic stories into which we write ourselves as protagonists: all of these have social and political histories that matter. We cannot adopt a viewpoint of no viewpoint', or pretend to a universality of view when our own view must necessarily always be only a partial one. The partiality of our analyses can be investigated not just in the terms of global politics and alternative human futures, but most simply in terms of the fact, stressed by de Beaugrande, that each of us. has only grasped a very limited subset of all the kinds of texts made on this planet in our lifetime. Indeed, as our societies grow more complex and internally differentiated, each of us arguably encounters an ever smaller fraction of the texts and the text-types of even one language. Here indeed there is special value in very large text corpora, where every sort of text made in the world is proportionally represented (an ideal not yet achieved of course in any real corpus). But even if we find such texts on the computer screen before us, what are we to make of them? We are not members of all these speech communities; we do not have available to us the intertexts considered by members to be relevant to their interpretation.

63

I believe that text linguistics must acknowledge the severe limitations of the heroic story of One Great Linguist who overcomes many obstacles to find the salvation of a Universal Truth about language. Those of us who seek to begin from diverse textual data and find genuinely useful typical patterns across large sets of texts may need to restrict our work, and our claims, to well-defined corpora with which we have adequate intertextual competence to work. If we seek generalizations beyond this limit, then we will need to work in teams or consortia, to pool our collective knowledge of texts in order to further our common knowledge of text. References Halliday, M.A.K. (1977). Text as semantic choice in social context. In T.A. van Dijk and J. Petfi, Eds., Grammars and Descriptions. Berlin : de Gruyter. Halliday, M.A.K. (1978). Language as social semiotic. London : Arnold . Lemke, J.L. (1983). Thematic analysis. Semiotic Inquiry 3(2):159-187. Lemke, J.L. (1985). Ideology, intertextuality, and the notion of register. In J.D. Benson and W. S. Greaves, Eds., Systemic Perspectives on Discourse. Norwood , NJ : Ablex Publishing, 275-294. Lemke, J.L. (1990). Talking Science: Language, Learning, and Values. Norwood , NJ : Ablex Publishing. Lemke, J.L. (1991). Text production and dynamic text semantics. In E. Ventola, Ed. Functional and Systemic Linguistics: Approaches and Uses. Berlin : Mouton/deGruyter, 23-38. Lemke, J.L. (1994). Intertextuality and text semantics. In M. Gregory and P. Fries, Eds. Discourse in Society: Functional Perspectives. Norwood , NJ : Ablex Lemke, J.L. (1998). Resources for attitudinal meaning: Evaluative orientations in text semantics. Functions of language 5(l): 33-56, 1998. Lemke, J.L. (in press). Semantic topography and textual meaning. To appear in J. de Villiers and R. Stainton, Eds. Communication in Linguistics. Toronto : Editions du GREF, 1-23

Sin and grace: naught for noughts? J.R. Martin University of Sydney Embarking on the noughts (2000-09), Robert and I are fellow-travellers. I agree the key to progress is reworking langue and parole as a dialectic of system and process, and finding ways of exploring process that illuminate metastability (systemic inertia and dynamism - the problem of social semiotic change). In relation to this project I find it useful to draw on two key dimensions of systemic functional linguistics 64

(hereafter SFL), the clines of instantiation and realisation (Halliday & Matthiessen 1999). Instantiation involves the way we observe metastability - as apparent flux or inertia or something in between. Hallidays analogy here is weather and climate; weather the capricious flux we experience day to day, climate the comforting inertia we try to use to plan. Critically, weather and climate are the same thing, looked at in different ways. And we can argue that weather changes climate, in ways that matter (global warming) and ways that dont (like when we hear that todays temperature is 26, 2 degrees above average, which in fact changes that average, and we dont fuss about the contradiction) -- or that climate determines weather (since It always rains in Melbourne, and as Sydney-siders overclaim). Thus text interacts with system, along a cline of instantiation, including system (the generalised meaning potential of a language), register (sub-potentials characterised as registers and genres), text type (generalised instances -- a set of texts if you will), text (the meanings afforded by an instance) -- to which we should perhaps add at the end of the cline, following Robert, reading (the meaning taken from a text according to the subjectivity of the reader): system (generalised meaning potential) register (semantic sub-potential) text type (generalised actual) text (affording instance) reading (subjectified meaning) Roberts point is that we cant understand the climate unless we study local weather. Amplifying this, Id suggest that we be clearer about how we position data and generalisations along this cline, and work harder at finding ways of modelling inertia and flux right along the cline - since we need to show both where the system end is leaning and how the reading end is constrained - in order to understand how the dialectic of instantiation changes things fast enough to matter and slowly enough to keep us sane. Realisation, on the other hand, involves the way we model metastable systems - as abstract or concrete, or something in between. Linguists, qua linguists, typically get real by taking phonic substance as point of departure, and treating the rest of language and culture as layers of abstraction. In SFL, depth of abstraction is treated as metaredundancy - as patterns of patterns of patterns of... (see Lemke 1995). Within language the strata focus on the syllable (phonology), clause (lexicogrammar) and text (discourse semantics); social context is modelled as field (institution), tenor (interaction) and mode (information flow), and sometimes in addition as genre (the system of social processes constraining immanent associations of field, mode and tenor relations; Martin 1992, 1997): genre (immanent social processes)

65

register (institution/interaction/information) discourse semantics (text forming resources) lexicogrammar (clause forming resources) phonology syllable forming resources/prosody) Instantiation and realisation are complementary dimensions in the model, so we can consider the relation of system to process at any level of abstraction. Robert didnt deal in detail with realisation, and what I want to do here is make some complementary suggestions about exploring language in use with this dimension in mind. By way of pointing us forward in our exploration of system and process, Robert draws on corpus linguistics, particularly its well automated dealings with collocation and colligation. I have no doubt that the collocational lexical perspective complements work on grammar as system in important ways, and that Hallidays notion of lexis as delicate grammar still has to be reconciled with this tradition. As for colligation, I expect we need to move from tagging syntagms to tagging structures (from class sequences like noun^noun or nominal^verbal to function frames like Actor^Process or Classifier^Thing) for significant progress to be made. The problem is of course that tagging functional frames is far harder to automate than tagging syntagms; we just dont have the parsers, and hand tagging is so slow. We have a soft-ware problem here, and we have to face it to make it go away. In the meantime, semi-automated workbenches will have to serve, as we move beyond classes to functions and thus bring meaning into the scope of grammatical inquiry, just as collocation research has brought lexical meaning to bear. So much for grammar; what about discourse? One thing that concerned me about Roberts exemplifications was that they didnt take us beyond the clause to deal with discourse semantic dependencies. In a journal like Text, this should perhaps be a central concern. Time to get our hands dirty. Consider text 1: [1] There is a sense, then, in which the demand for a total recollection of sins results in the unlimited extensions of discourse purporting to extract and convey ones successes and failures in accounting for past acts and desires. Accounting thus allows confession to become a self-sustaining machine for the reproduction not only of Gods gifts of mercy but of sin as well. For Gods continued patronage -- the signs of His mercy -- requires a narrative of sins to act upon. The confessor who sits in lieu of an absent Father needs the penitents stories, without which there can be no possibility of asserting and reasserting the economy of divine mercy. Without the lure of sin, the structure of authority implicit in this economy would never emerge. Confession was crucial because it produced a divided subject who was then made to internalise the Laws language. The penitent became the speaking subject who is also the subject of the statement (Foucault 1980: 1:61). But confession was also important because it made for the 66

ceaseless multiplication of narratives of sin through their ever-faulty accounting. In introducing the category of sin, confession converted the past into a discourse that was bound to the Law and its agents. In this way the accounting and recounting of the past generated the complicitous movement between sin and grace. [Rafael 1988] Im sure collocation and colligation would have a lot to say about data of this kind, whether working with a span of 80 or 200 characters or more. The lexis and grammar of post-colonial discourse feels distinctive; and we could begin to systematise our feelings. But there is more going on. Rhetorical dependencies between sentences and ranking clauses for one thing, as the argument unfolds: [i] There is a sense, then ... [ii] Accounting thus ... [iii] For ... [iv] [v] [vi] Confession was crucial because it produced a divided subject who was then ... [vii] [viii] But confession was also important because ... [ix] In introducing the category of sin... [x] In this way ... And identity dependencies as participants are tracked: Gods-Gods-His Confession-it confession-it narratives-their the Law-its And these two types of semantic interdependency interact at both the beginning and end of the text. At its beginning, resolving the identity of the demand for a total recollection of sins specifies the scope of the linker then, which on its own simply tells us that what preceded is causally connected to what follows (taking us back to The Spanish demand is that nothing be held back in confession; see text [1] below). [The Spanish demand is that nothing be held back in confession] 67

There is a sense, then, in which the demand for a total recollection of sins Similarly at the end of the text, resolving the identity of the accounting the recounting of the past specifies the scope of this in the linker in this way. Exploring further, for example through the taxonomy oriented ideational dependencies (noting in particular the balance of semiotic and religious lexis)... narrative-stories-narratives-accounting-accounting-recounting demand-asserting-reasserting-speaking sins-successes-failures-mercy-sin-mercy-sins-mercy-sin-sin-sin-sin-grace God-God-confessor-Father-penitent-penitent etc. we should arrive at a schema naturalised by the text in which [i-iii] are elaborated by [iv-ix] which are in turn elaborated by [x]; the function of confession is previewed, expanded upon, then distilled as the complicitous movement between sin and grace: [i] There is a sense, then, in which the demand for a total recollection of sins results in the unlimited extensions of discourse purporting to extract and convey ones successes and failures in accounting for past acts and desires. [ii] Accounting thus allows confession to become a self-sustaining machine for the reproduction not only of Gods gifts of mercy but of sin as well. [iii] For Gods continued patronage -- the signs of His mercy -- requires a narrative of sins to act upon. [iv] The confessor who sits in lieu of an absent Father needs the penitents stories, without which there can be no possibility of asserting and reasserting the economy of divine mercy. [v] Without the lure of sin, the structure of authority implicit in this economy would never emerge. [vi] Confession was crucial because it produced a divided subject who was then made to internalise the Laws language. [vii] The penitent became the speaking subject who is also the subject of the statement (Foucault 1980: 1:61). [viii] But confession was also important because it made for the ceaseless multiplication of narratives of sin through their ever-faulty accounting. [ix] In introducing the category of sin, confession converted the past into a discourse that was bound to the Law and its agents. [x] In this way the accounting and recounting of the past generated the complicitous movement between 68

sin and grace. And this is just one small piece of texture in a longer phase the third part of Chapter 3 of Rafaels Contracting Colonialism, a treatise on translation and Christian conversion in the Philippines : 3. Conversion and the Demands of Confession 84 The inadequacies of Tagalog Conversion 84 Reducing Native Bodies 87 Confession and the Logic of Conversion 91 This longer phase opens with a transition from what preceded (this internalisation...) to what ensues (two interrelated procedures); these procedures are scaffolded as they appear by the linker first, and the phoric Numerative second - and resolved through the identity of the process of accounting and the discourse of interrogation. This internalisation of an exterior hierarchy consists of TWO interrelated procedures: the accounting of past events and the reproduction of the discourse of interrogation contained in the confession manual: First, the process of accounting. All confession manuals contain the unconditional demand that all sins be revealed... These considerations bring us to the second moment in the interiorisation of hierarchy prescribed by confession: the reproduction of the discourse of interrogation... Many people find post-structuralist discourse hard to read. And it is one of the most abstract discourses the technology of writing has enabled writers to evolve. But its rhetoric and the way it is textured through semantic dependencies of the kinds just reviewed subsumes the more familiar rhetoric of modernist discourse (Halliday 1998, Wignell 1998). The challenge for text analysts is to unveil this rhetoric, and explain just how the post-colonial discourse has superseded it. We didnt get far down that road here. Heres the text again, with some co-text; if knowledge about language and social context has a role to play in language learning, it should be easier going this time round. [1] This internalisation of an exterior hierarchy consists of two interrelated procedures: the accounting of past events and the reproduction of the discourse of interrogation contained in the confession manuals. First, the process of accounting. All confession manuals contain the unconditional demand that all sins be revealed... The Spanish demand is that nothing be held back in confession. One is to expend all that memory can hold in a discourse that will bring together both the self that recalls and that which is recalled. The present self that confronts the priest in confession is thus expected to have managed to control his or her

69

past - to reduce it, as it were, to discursive submission. Whereas the examination of conscience requires the division of the self into one that knows the Law and seeks out the other self that deviates from it, a good confession insists on the presentation of a self in total control of its past. It is in this sense that confessional discourse imposes on the individual penitent what Roland Barthes called a totalitarian economy involving the complete recuperation and submission of the past to the present, and by extension of the penitent to the priest (Barthes 1976: 39-75). Yet insofar as the ideal of a perfect accounting of sins also necessitated their recounting in a narrative, it was condemned to become a potentially infinite task. Given the limitations of memory, accounting engenders its own errors. And the errors created by faulty accounting become further sins that have to be added to the original list. The very possibility of a correct accounting engenders an erroneous accounting, just as remembering ones sins would make no sense unless there existed the possibility of forgetting them. It is thus the guarantee of a faulty accounting of sins that makes conceivable the imperative for total recall. Barthes puts it more succinctly: Accountancy has a mechanical advantage: for being the language of a language, it is able to support an infinite circularity of errors and of their accounting (Barthes 1976: 70). There is a sense, then, in which the demand for a total recollection of sins results in the unlimited extensions of discourse purporting to extract and convey ones successes and failures in accounting for past acts and desires. Accounting thus allows confession to become a self-sustaining machine for the reproduction not only of Gods gifts of mercy but of sin as well. For Gods continued patronage -- the signs of His mercy -- requires a narrative of sins to act upon. The confessor who sits in lieu of an absent Father needs the penitents stories, without which there can be no possibility of asserting and reasserting the economy of divine mercy. Without the lure of sin, the structure of authority implicit in this economy would never emerge. Confession was crucial because it produced a divided subject who was then made to internalise the Laws language. The penitent became the speaking subject who is also the subject of the statement (Foucault 1980: 1:61). But confession was also important because it made for the ceaseless multiplication of narratives of sin through their ever-faulty accounting. In introducing the category of sin, confession converted the past into a discourse that was bound to the Law and its agents. In this way the accounting and recounting of the past generated the complicitous movement between sin and grace. [Rafael 1988: 101-103] These considerations bring us to the second moment in the interiorisation of hierarchy prescribed by confession: the reproduction of the discourse of interrogation... My point here is that we need discourse semantic tagging, alongside lexicogrammatical analysis, in order to unpack the rhetorical contingencies whereby texts make meaning -- including the meanings that make and re-make system. There is more to system than grammar;[Note. Even where discourse semantics and phonology are given a place in

70

analysis, promotion of grammar as the semogenic powerhouse of the system gives me pause (cf. Halliday & Matthiessen 1999) -- discourse analysts might not agree.] it involves phonology/graphology and discourse semantics as well. Collocation and colligation, however richly conceived, will never tell us all we need to know. However much harder the tagging of discourse semantic dependencies may be to automate, I think we are in desperate need of more, systematic text analysis that goes beyond the clause without lapsing into informal explication de texte. Martin (1992) suggests four major regions of discourse semantic analysis - identification (participant tracking), conjunction (logical connections of time, cause etc.), negotiation (speech function and dialogue structure) and ideation (realisation of taxonomies and activity sequences). It strikes me that research in SFL and west-coast functionalist tradition[Note. I take Fox 1987 as the exemplary west-coast functionalist study, since it brings several discourse semantic regions (CA, RST and participant identification) to bear on the grammar of text development; more generally, dialogue with SFL has been impeded by west-coast aversion to theory building and their concern with discourse as an explanation for grammar, at the expense of developing grammars as (part of) an explanation for discourse (cf. Cumming & Ono 1997).] has been converging around these regions for some years and that productive dialogue is now possible. Im thinking here for example of Dubois 1980, Mann & Thompson 1992 in relation to Halliday & Hasan 1976, Martin 1992; of Ochs et al 1996 in relation to Coulthard 1992; of Halliday & Matthiessen 1999 in relation to Langacker 1987: 'West Coast Functionalism' (Fox 1987) identification identity (Dubois 1980) conjunction negotiation ideation RST (Mann & Thompson 1992) CA (Ochs et al. 1996) cognitive linguistics (Langacker) SFL reference conjunction exchange ideational semantics

No doubt readers can make additional connections of their own. The critical thing is that discourse analysis include text analysis, not as a matter of form, but as the semantic foundation for discourse on discourse -- the meta-readings were trained to make. The cline of abstraction I introduced above includes the social -- as a pattern of meanings. As such it makes room for a language-based theory of social context, which enables linguists to participate in transdisciplinary projects as social semiotic practitioners. I think the time for productive interdisciplinary work is over -- by which I mean projects in which linguists hand over to sociologists, anthropologists, critical theorists or whatever once theyve worked through phonology, lexicogrammar and discourse semantics (or even before!). This doesnt encourage dialogue, since it means the linguists arent taking responsibility for the social just as the social theorists arent taking responsibility for 71

language (cf. Schegloff 1996). We need overlapping intrusive expertise to move on,[Note. What WILL happen is that well continue to be bogged down in common sense chat about social context- - at times referred to as ethnography (but not informed that I can see by social theory of any kind), at times reglossed as cognition (but not reconciled that I can see with Edelmans neo-Darwinian neuro-biology; cf. Halliday 1995, Halliday & Matthiessen 1999, Matthiessen 1998 for discussion).] and this means pushing the cline of realisation right through as many levels of abstraction as we can. Martin (1992) interprets field, tenor and mode (named register) as configurations of ideational, interpersonal and textual meaning respectively; and interprets genre as co-patterning of register choices recurrently phased together in unfolding text. Genre in such terms, as a pattern (of a pattern of) linguistic choices corresponds to Bibers use of the term text type (cf. Biber & Finnegan 1995: 7-10). Bibers reserves the term genre (later register) for folk categorisations of discourse glossed in terms of social purpose, and packages his corpora for both analysis and interpretation in relation to such criteria, which he sees as language external. As dialogue between SFL and this great Northern Arizona tradition of corpus based research unfolds I would like to see a richer tagging system, including function structures and discourse semantic dependencies, developing in tandem with an ongoing re-packaging of corpora based on genre and register as configurations of meaning. Initially, of course, the theory behind such packaging will depend on register and genre theory evolving out of intensive manual analysis of exemplary texts (Christie & Martin 1997); as automation facilitates analysis, this can be extended along the cline of instantiation through text types to registers. In the foreseeable future the need for Bibers distinction between text type and register/genre would hopefully disappear, as we arrive at a linguistically responsible characterisation of language use as recurrent configurations of meaning. At this point linguistics will have arrived as a real player in the humanities and social sciences, with a linguistically materialised theory of social action -- and the transdisciplinary dialogues we need can take off. But without richer tagging, and corpora packaged with respect to current best guesses about immanent genres, this isnt going to happen in the noughts.[Note. In Australia, for example its the teacher/linguists who have pushed our literacy work ahead on a transdisciplinary footing, not linguists working with teachers; Cope & Kalantzis 1993, Hasan & Williams 1996, Christie 1999.] In short, Im agreeing with Robert about instantiation -- system and process in relation to genesis is what were after; and Im expanding on Robert in relation to realisation - we need to get beyond the clause, getting bigger and digging deeper towards a fuller spectrum of social linguistic analysis.[Note. Its no use getting deeper without getting bigger, along the dated syntax, semantics and pragmatics cline, since in this tradition we never really escape the clause; text analysis gets pushed to the margins, as performance really, as Robert implies.] Pursuing this, we have to be cautious of two things: (i) getting trapped by automation, so that we only do what machines let us (e.g. collocation, colligation,

72

text types based on words classes and syntagms etc.); (ii) getting mesmerised by scintillating grammarians proffering super-grammars, so that we put off discourse analysis because the super-grammars do so much more than we could ever have reasonably expected them to. Believe me when I say that Im not slighting Biber, Sinclair, Halliday and Matthiessen here; their work founds our future, and grammars like Hallidays, however rich from a grammarians point of view, are barely enough when it comes to serving as one key meta-semiotic ratchet in the discourse analysts toolkit. But as discourse analysts we have to put ourselves in position to bargain strongly with both grammarians and soft-ware. To grammarians were saying, Fine; give us all you got; now, give us more; and by the way, when you run out of grammar, let us take over. To programmers were saying, OK; give us an interactive workbench for rich text analysis; automate what you can, and well do the rest by hand; and by the way, please build a program than can learn from our manual analysis and from our manual editing of your automations how to automate better and automate more! If we bargain well, we can move forward gracefully, wary of sins of omission (however technologically induced). Bargain badly and it will be naught for noughts. Robert to umpire. Us to choose. References Biber, D. & E. Finnegan. 1994. Sociolinguistic Perspectives on Register. Oxford : Oxford University press. Coulthard, M. [Ed.] .1992. Advances in Spoken Discourse Analysis. London : Routledge. Christie, F. [Ed.]. 1999. Pedagogy and the Shaping of Consciousness: linguistic and social processes. London : Cassell. Christie, F. & J. R. Martin. 1997. Genre and institutions: social processes in the workplace and school. London : Cassell (Open Linguistics Series). Cope, W. & M. Kalantzis [Eds.]. 1993.The Powers of Literacy: a genre approach to teaching literacy. London : Falmer. Cumming, S. & T. Ono. 199.7 Discourse and grammar. T A van Dijk [Ed.] Discourse as Structure and Process. London : Sage (Discourse Studies: a multidisciplinary introduction. Volume 1).112-137. Du Bois, J. W. 1980. Beyond definiteness: the trace of identity in discourse. W L Chafe [Ed.] The Pear Stories: cognitive, cultural and linguistic aspects of narrative production. Norwood , N.J. : Ablex. 203274. Fox, B. 1987. Discourse Structure and Anaphora: written and conversational English. Cambridge : Cambridge University Press. 73

Halliday, M. A. K. 1995. On language in relation to the evolution of human consciousness. In S. Alln [Ed.] Of Thoughts and Words: Proceedings of Nobel Symposium 92, The relation between language and the mind, Stockholm 8-12 August 1994. London : Imperial College Press. Halliday, M. A. K. 1998. Things and relations: regrammaticising experience as technical knowledge. In Martin & Veel 185-235. Halliday, M. A. K. & R. Hasan. 1976. Cohesion in English. London : Longman (English Language Series 9). Halliday, M. A. K. & C. M. I. M. Matthiessen 1999. Construing Experience through Meaning: a language-based approach to cognition. London : Cassell. Hasan, R. & G. Williams [Eds.]. 1996. Literacy in Society. London : Longman. Hunston, S. & G. Thompson [Eds.]. 1999. Evaluation in Text: authorial stance and the construction of discourse. Oxford : Oxford University Press. Langacker, R. 1987. Foundations of Cognitive Grammar. Stanford: Stanford University Press. Lemke, J. 1995. Textual Politics: discourse and social dynamics. London : Taylor & Francis. Mann, W. C. & S. Thompson [Eds.] 1992. Discourse Description: diverse analyses of a fund raising text. Amsterdam : Benjamins. Martin, J. R. 1992. English Text: system and structure. Amsterdam : Benjamins. Martin, J. R. & R. Veel 1998. Reading Science: critical and functional perspectives on discourses of science. London : Routledge. Matthiessen, C.M.I.M. 1995. Lexicogrammatical cartography: English systems. Tokyo : International Language Sciences Publishers. Matthiessen, C.M.I.M. 1998. Construing processes of consciousness: from the common sense model to the uncommon sense model of cognitive science. In Martin & Veel, 327-356. Ochs, E., E. A. Schegloff & S. A Thompson. [Eds.]. 1996. Interaction and Grammar. Cambridge : Cambridge University Press. Rafael, V. 1988. Contracting Colonialism: translation and Christian conversion in Tagalog society under early Spanish rule. Manila : Ateneo de Manila University Press, 101-103. Schegloff, E. A. 1996. Turn organisation: one intersection of grammar and interaction. In Ochs et al. 52133. Wignell, P. 1998. Technicality and abstraction in social science. In Martin & Veel 297-326.

74

Last words? Robert de Beaugrande These responses certainly do satisfy the criteria of range and diversity, which I of course find gratifying but which also doom any attempt at riposte or rebuttal within the confines of an already lengthy, erm, intertext. I confess being a trifle disquieted when ideas or positions I do not hold, and have said why not in print, albeit perhaps not here, are attributed to me now a reflex I suppose of unfamiliarity with the larger scope of my work. But I am quite content to rest my case with what I said here and now. Still, just one example: I have always rejected the notions of the text being a unit above the sentence or a sequence of sentences (pace Harweg): the text is an event, (yes, an actualisation!) whereas the sentence is at most the default unit (also actual!) of written language as opposed to the utterance of spoken language. I imply no opposition of sentence and text, so I cannot have abandoned it (Harweg again). Since I had only the responses e-mailed to me, I have taken the liberty of unifying the references as well as insidious formats stemming from disagreeable word-processors which padded in assyrian hosts of unwanted blanks or confounded me with unsearchable phantom menaces wreaking baleful and hideous side-effects upon prints and paragraphs. More prosaically, I amended obvious typos and putative lapses in punctuation, but only the more distended or bewildering Teutonic solecisms. All in all, I am delighted to finally present this colourful gallery, and to thank my distinguished colleagues for their thought-provoking deliberations.

75

76