AW: Widdowsons book analysed

Wodak, Ruth r.wodak at LANCASTER.AC.UK
Thu Mar 9 13:34:44 UTC 2006

Very good review, thanks! Another review written by myself will appear in Language in Society in the fall.
Ruth Wodak
Professor in Discourse Studies
Department of Linguistics and English Language
Lancaster University
Lancaster, UK LA1 4YT
Tel: xx44 1524 592437
Fax:xx44 1524 843085


Von: Critical Discourse/Language/Communication Analysis im Auftrag von bea2gran at IDECNET.COM
Gesendet: Do 09.03.2006 11:35
Betreff: Widdowsons book analysed


The Case against Critical Discourse Analysis Reopened:

In Search of Widdowson's "Pretexts"


Robert de Beaugrande

Università del Litorale

Koper - Capodistria


Imagine if you will a most peculiar game of bowling. One contestant is the sole bowler; the others are all human bowling pins. And instead of an array of pins in the centre of the "alley", just one human pin at a time is set up in the "gutter" where the bowler cannot fail to knock it down. And so he does, on and on, with evident relish in his underpinned mastery.

Such a game seems to me a not altogether unfitting metaphor for H.G. Widdowson's (2004) latest and largest - and, one ardently hopes, last - assault on "(Critical) Discourse Analysis". Some of the targets are familiar from his "past ten years" (page x) of fulminations, which he esteemed highly enough to have collected in a special volume, Controversies in Applied Linguistics (Seidlhofer [ed.], 2003); and he frankly calls the new book "confrontational and uncompromising" ("waspish" would have been my word). Indeed it is; yet he does not seem deterred by the prospect that such a stance just might affect his own "rationality" and "logical consistency" (see below). The Preface demurely trivialises his verbal sallies of aggression, which he prefers to call "reservations" and the like: "no offence is intended, and I hope to be forgiven if any is taken" (ix). Yet much of it is hard for me to see as inoffensive; and the colleagues he attack and I have corresponded with did not sound exactly forgiving. Here is what just one of them e-mailed me after reading a draft of this piece:


[1] When he attacked me without warning in Georgetown he used as his text a short scrap I wrote up from a discussion at a British Council conference, rather than the other things, such as a book-length defence of the use of corpus evidence, which I had published more recently. (John Sinclair)


And that's not all he wrote. Responses from other prominent victims were, erm, considerably less adaptable to public citation.

 In personal discussions and later in written ripostes (e.g Beaugrande 1998a, 2001 - mysteriously missing from Controversies), I have for years attempted to persuade Widdowson that his mission as a personage "renowned in the fields of applied linguistics and language teaching" (rear book cover blurb) should not be to harp on judging and dismissing alternative methods of language inquiry, but to go on and provide his own full-fledged theory or method to achieve what he complains they do not. When I heard he had signed a contract for a book on discourse analysis, such, I hoped, was his design. But the present book seems to leave to readers the formidable task of cobbling together his own method from a diffuse and often apodictic gallery of caveats of how not to proceed; and so few of the most eminent workers in discourse analysis and various related field remains unscathed or at least unslapped that one wonders what on earth is left for approval.

Evidently without affecting him, I have deconstructed his criticisms of Halliday, Sinclair, and Fairclough and their three respective fields of inquiry (Beaugrande 1998a, 2001). Curiously, I am (or was) a friend (and sometime boozing buddy) of all four gentlemen, though of course never at the same time and place; in a relaxed environment, they are genial chappies you wouldn't imagine in combat. For myself, I could see the point and viewpoint of each one; I didn't see any reason why they shouldn't go their separate ways in peace. Besides, I am practised a long-term synthesizer of work by broad fields and researchers (e.g. Beaugrande 1980, 1984), some of whom were not even on speaking terms with each other nor allowed any merit to each other's approaches. In my earlier years on the active conference circuit, I would at times be regaled in mutual succession with harsh reproofs by one against another, whilst I sat tranquilly with what I hoped was an owlish and inscrutable countenance; yet I remained determined to judge for myself on each one's own merits. The razor of my own polemics were aimed precisely at those who revelled in unfair or over-the-top polemics, as when Chomsky dismissed other academcis' work as 'uninteresting', 'unscientific', 'obscure', 'unserious', 'puerile', 'banal', 'unsophisticated', 'gross', 'careerist', 'propagandist', 'pretentious', 'dogmatic', 'distortive', 'irrational', 'immoral', and 'vile' (cf. Beaugrande 1999b for his sources and my right and proper send-up).

So I was, so to speak, automatically activated by Widdowson's (2000) polemics against Halliday, Sinclair, and Fairclough, and undertook to show, via "discourse analysis" of a mini-corpus" Widdowson's diatribes, using actual quotes from all three scholars, that he was manifestly uninformed, misguided, or unfair. These papers freely available are on my website at, and I need not unroll the lines of argument here again. I merely note with minor relief that Sinclair, is no longer a major Aunt Sally; yet in return, his list of supposedly misguided discourse-analysing "academics" of all cuts and stripes has mushroomed like a cluster-bomb.

In his own vision, this new "book is, in a sense, a reconceptualised and extended version of one that was unwritten thirty years ago", namely the "write-up" of his "PhD thesis" (Widdowson 1973) - a strange "sense" it must be, since most of the works he attacks with a ferocity that increases as they grow more recent and popular, were published after 1973. He now "immodestly acknowledges" that this hushed and darkling tome of his salad days was the "first" of all that "addressed and tentatively explored" "many of the issues in discourse analysis". True, it was based on the bare-bones and early structuralist "discourse analysis", limited to a handful of sentences, by Z.S. Harris (1952), who immediately and permanently abandoned it for a "transformational" approach (e.g. Harris 1957), which was promptly expropriated by his pupil Chomsky (1957) for a larger scheme with a smaller unit ­ - the "sentence", casually handled as either theoretical or practical unit. Harris edified us with the "analysis" that the phrase "satisfied customers" on a bottle of hair-tonic can be restructured as "customers will be satisfied", though I can't resist imagining that on the door of a massage parlour instead. This same Harris is now resurrected as a principal witness in the massive show-trial of recent "discourse analysts" and their kindred, apparently implying that "discourse analysis" jolly well ought to have pursued his lines. For the record, I recalls, that he seriously proposed to exclude "meaning" from a purely "formal" study of language (Harris 1951); yet even he couldn't manage that method, and used "meaning" on the hand-waving reassurance that it would arrive at the same results as a "purely formal" method.

When "discourse analysis", became fashionable, Widdowson now owns, he was "chagrined" that his own early efforts went unnoticed" "without a nod of recognition", though how one "nods" at "unwritten" work is a trifle hard to imagine. At all events, the 1973 "book" (i.e. thesis) is now offered for free download on the Internet, with unwonted generosity, by Oxford University Press and is on my master home computer. To call the new book a "reconceptualised and extended version" is a master stroke of British understatement.  

Still harder for me to imagine is how the "address and explore" of 1973 can have transmogrified to "indignantly rebuke and deplore", which I find a better description of what the 2003 write-up mostly does or tries to do. The book bristles with blanket dismissals in bold and broad terms, viz.:


[2] We are left with nothing that is secure enough to get our bearings from. It is difficult to see how CDA [Critical Discourse Analysis] [...] can be based on any secure guiding principles at all. (168)

[3] There is no grappling here with intellectual uncertainties, no confrontation of opposing paradigms. Nothing comes across as posing any real problems. (168)


Having myself published at least three thick volumes (also posted for free on the Internet and nowhere mentioned by Widdowson) that utterly belie these reproofs (Beaugrande 1991, 1997, 2004) I am frankly gobsmacked. I have presented resolutely secured and documented guiding principles by working upwards toward communication and discourse via the evolution of complexity and information through mathematics, physics, chemistry, biology, neurology, anthropology, ethnography, and sociology, e.g., in the general linear principles that can operate complex systems and their interactions. Using only the tools of discourse analysis, I have grappled with such intellectual uncertainties as enshroud the Chapman-Kolmogorov equation and the Belusov-Zhabotinsky reaction as candidate models of linear-life-systems capable of information, and resolved, I believe, the superiority of the latter; I have confronted opposing paradigms inside and outside linguistics; and I have always insisted that the hardest problems for discourse analysis are just beginning to reach full force. And for the record, my new book (2004, Ch. III) applies a Halliday-style lexicogrammar to exactly 701 samples of authentic texts, which Widdowson, citing mostly just the Preface of Halliday's (1994) Introduction, maintains cannot be done at all; Halliday (1973) himself famously did so for Golding's The Inheritors, which was a seminal inspiration piece for many, myself included.

Liberally interspersed in Widdowson's one-man bowling-game are assertions that CDA is not "analysis", and doesn't properly grasp what "discourse" is anyhow. He departs from a distinction that does taste rather like 1973, between text which can be analysed versus discourse which must be interpreted. In the early 1970s, linguistics was radically theoretical - arcane, formalist, and top-heavy - but reverently devoted to the "sentence". So when the "text" was grudgingly admitted, it was construed in some influential work (e.g. van Dijk 1972) as the theoretical unit, and "discourse" as the practical unit. Perhaps the dichotomies of "langue" versus "parole" and "competence" versus "performance" helped along as contumacious mandates that "language" can be "scientifically analysed" only by theorising, abstracting, idealising - in short, behaving as little as possible like ordinary participants in contexts of communication. But in my view, the outcome was merely a bizarre charade of camouflaged communication where the lion's share of the analyst's work was kept out of sight by invoking the "native speaker's intuition" and "introspection" (cf. Beaugrande 1998b). For the new Widdowson of 2003, "discourse analysis" must move beyond the "interpretation" "engaged in" by "ordinary interpreters", by "investigating" "how" they do it, but it dismally fails (155). 

By the late 1970s, at all events, most large-scale projects to "theorise the text" (of which Widdowson seems unaware and which were mortally unreadable for me back then) had collapsed under their own weight in a unending morass of "rules" and "features" (e.g. van Dijk, Ihwe, Petöfi, and Rieser 1972), some numbering in the thousands. And, I maintained, for the simple but seemingly invisible reason that the text is more productively conceptualised as a practical unit intended as a contribution to a discourse (Beaugrande and Dressler 1981). Widdowson now seems to animate his own division: the "text" not so much as a theoretical unit hovering in some abstract nirvana of "competence" as an analysable unit according to a "rigorous theory", which I expected to discover from his book but found myself unable to do so. To be sure, especially in print, the text beguiles us as pure form with its seeming self-evidence, familiarity, and simplicity because we approach it by immediately and unconsciously performing what Widdowson calls "interpretation", which his book at times construes as some mode of intervention, commitment, even ideology - a "distorting influence" that "shapes and colours  analysis", as I said (155) - in short, the work of a "pretext" that compromises "analysis" along with "academic objectivity" and "rigour".

The rebuke is like that bowling pin knocked down in the gutter insofar as CDA workers like myself, van Dijk, and Fairclough (who is knocked down and reset the oftenest) regard "objectivity" and "rigour" themselves as real (but well-hidden) problems that can legitimise academic analysts in presenting "analyses" purified of social and political positioning - which may be fine for femtosecond technology using ultra-fast-pulse lasers to generate 'photo-dissociation' (the reverse of the celebrated 'photo-synthesis'), but surely not for discourse. My latest book is emphatically clear on this point and supported with plentiful authentic discourse data (Beaugrande 2004), if 2,382 data samples may justly be called "plentiful". The ruthless global consolidation of power, wealth, and the media in the hands far-right "conglomerates" - some optionally disguised as "governments"- has been accompanied by a radical recentralisation or discursive evasion. Compare and contrast [4] with [5], and [6] with [7]:


[4] Since taking office, the Bush Administration has stopped work on dozens of important safety and health standards, withdrawn worker training grants and stopped important record-keeping rules that would require employers to identify which injuries are musculo-skeletal disorders. (AFL-CIO)

[5] Exposure to ergonomics-related injuries is not well-understood or easily measured, making regulations for all industries difficult. (US "Labor" Department).

[6] Eugene Scalia [nominated for "Solicitor of the Department of Labor" and the son of the "Supreme Court Justice" who helped inflict a "President Bush" on the world] refers to repetitive-stress injuries, which afflict 600,000 American workers annually, as "junk science", [and] a "psychosocial issue" - in effect, calling those who suffer from it fakers [...] "who do not like their jobs." (Joshua Green in American Prospects)WWW

[7] The administration provides no proof or credible argument that the proposed rule [that the United Parcel Service pay for protective equipment such as respirators and gloves] will improve health and safety, and in fact, the rule will cause significant economic harm, will not promote health and safety, and may reduce personal protective equipment by reducing collectively bargained cooperation between union and management in the implementation of personal protective equipment requirements. (Scalia's own brief)


By this duplicitous discursive logic in [7], the obvious fact that gloves provide safety still needs to be "argued" and "proven"; company expenses constitute "economic harm"; and mandating "protective equipment" equals "reducing" safety by stirring up antagonism between "union and management", which would of course actually result from not getting it. Or, consider these:


[8] "Junk science" is the term that corporate defenders apply to any research, no matter  how rigorous, that justifies regulations to protect the environment and public health; [...] "sound science" is used in reference to any research, no matter how flawed, that can be used to challenge, defeat, or reverse environmental and public health protections (Rampton and Stauber 2001: 126f, 222f). 

[9] "Junkman" Steven Milloy has made a career of lobbying for polluting industries, heading corporate front groups to deny environmental concerns, and ridiculing individual environmentalists. [...] Milloy defines "junk science" as "bad science used by lawsuit-happy trial lawyers, the "food police", environmental Chicken Littles, power-drunk regulators, and unethical-to-dishonest scientists to fuel specious lawsuits, wacky social and political agendas, and the quest for personal fame and fortune". (Boston 1998: internet)


Never mind that it's Rent-A-Rants like Milloy who get paid a "fortune".

These are the sorts of discursive issues that CDA seeks to bring to light and confront without our own "counter-discourse", and therefore sees slight value, if not actual handicap, in the sort of academic window dressing Widdowson's rebukes us for lacking. He recognises this objection, but patronisingly enlightens all us ignoramuses that "academics engage in intellectual enquiry and do research", whereas "the promotion of the critical cause by persuasive appeal at the expense of academic rigour" "does the cause a serious disservice" (173).

If there is a really "serious disservice" involved, then it is the obvious failing of language study to promote and support critical awareness of discourse. But Widdowson seems to have some foreknowledge of that too, though only in the Preface - by far the book's biggest surprise:


[10] There has surely never been a time when the need for such an investigation is so urgent, when public uses of language have been so monopolised to further capitalist interests to the detriment of well-being and in denial of human rights and social justice Ecological devastation goes under the verbal guise of economic development, and millions of people are kept subject to poverty, reduced to desperation, deprived of liberty and life in the name of democratic values and a globalized market economy that is said be free. (viii)


Before readers can recover their breath, he vows he "wholeheartedly endorses" "the cause of CDA", and actually thanks his gallery of bowling pins for "indispensable contributions to this book" (ix) These evidently consist of "pretexts" (in my sense, not his) for "having reservations" and "calling into question" specific spots and fuzzy bit of discourse analysis - say quarrels over the "meaning" of a single word or short phrase - which, I admit, are in the main unsatisfactory and plausibly cherry-picked to project a feeble image of CDA. For example, the newspaper headline (source not given)




is said to be "an obvious intertextual echo of Shakespeare's Hamlet", whereupon Widdowson seems to charge Fairclough with failing to concede the "innumerable cases where it is impossible to decide whether a sequence of words" "is a snatch of another" (148). Why this factor, in the gross and scope of Widdowson's opinion, should bode some strange eruption to Critical Discourse Analysis is not elaborated.

For CDA, the real point of interest is how HRH Camilla's-Boy said it, which Widdowson did not quote, much less analyse, so here it is:


[12] The Prince of Wales [...] declared the English language had declined into a "dismal wasteland of banality, cliché and casual obscenity". [He] deplored "the abandonment of learning the rules of grammar by rote" and stressed that higher standards of literacy were needed if Britain was to compete in an increasingly competitive world. (Guardian, 1989)


In this royal rebuke, the language teaching profession is implicitly made one scapegoat for Britain's inability "to compete", as if the sagging economy were a matter of rules of grammar. Brian Cox of Black Paper fame suggests in his Cox on Cox an even more heinous scapegoating within Tory ideology: 


[13] Norman Tebbit, later Chairman of the Conservative Party, claimed that the decline in the teaching of grammar had led directly to the rise in football hooliganism. Correct grammar was seen by him as part of the structures of authority, such as respect for elders, for standards of cleanliness, for discipline in schools...


So language teachers are also indirect perpetrators of social violence. I could relish nothing more than to see what would ensue if Stormin' Norman, later Lord [sic] Tebbit in his natty suit, made a guest appearance before "football hooligans" to administer a ceremonious lesson on "correct grammar". 

But Widdowson's book shies away from data bearing on these large social issues, despite the bold avowal in [10] - like nailing one's colours to the mast of the ship and then disembarking before it sails. So his book turn out as his own rather miscellaneous and rambling "critical analysis" (or more precisely "meta-analysis") of particular flaws in very different (C)DA projects and methods, like a laundry-list of clothes hung up on a public clothesline according  to how each individual item seems torn or stained.

To deconstruct these captious quarrels one by one would require a book as large as Widdowson's, and, in fact, a deal more objective, rigorous, and organised He seems to imagine that he can simply dispose of discourse analysis done by established methods, especially critical discourse analysis, plus systemic functional grammar, sociolinguistics, speech-act theory, and much else, making them all just collapse and disappear, or at least quail sheepish and embarrassed about their procedures and results under the "critical" scrutiny of his Jeremiads ­- which I suspected might gladly be threnodies. ("Woe unto their soul! The shew of their countenance doth witness against them!"  Isiah 3:9) 

Meanwhile, bleary-eyed readers might ask with mounting frustration: if all these would-be "discourse analysts" have got it wrong, when is Widdowson going to present his own method that sets matters aright? This does not expressly occur, as far as I can see, until pages 169-171 (the book has x+ 174 pages):


[14] We can use the term "discourse analysis" to refer to the process of enquiring into textual facts and contextual and pretextual factors acting upon each other in the interpretive process. [...]

[15] One way of proceeding might be to establish default interpretations of text based on psycholinguistic research [and] by postulating an idealised lay reader [a cousin of Chomsky's "ideal speaker-hearer"?]

[16] Ethnographic inquiries might be carried out into how readers of different social cultural background and political persuasions actually respond to texts of various kinds

[17] One way of proceeding would be to elicit the reactions of readers of the original text and a version of it in which the linguistic features [...] have been systematically changed.


Only the "procedure of retextualisation" (which had been proposed before but mainly against the practices of discourse analysts like Wodak, 142-43) - a tactic much more extensively and intriguingly deployed with poetry in Widdowson's Practical Stylistics (1992) - is demonstrated on discourse data, and with cryptic briefness (and source unnamed);


[18] The essential aim of antenatal care is to ensure that you go through pregnancy and labour in the peak of condition [...]

[18a] The essential aim of antenatal care is to ensure that women go through pregnancy and labour in the peak of condition [...]


How this minute change would "give rise to effects" is not even tersely suggested; it seems to me, either way is a smarmy official counsel for a "pregnant woman"; to my knowledge, pregnancy is not (yet) general feasible in Britain for a man, though many volunteered when the prospect was raised; still, we might get some more visible "effects" if we changed it to "anti-natal care" if someone thought that might mean "abortion". Widdowson merely recommends us to try "empirical investigation"; none is cited, though the research literature was already quite substantial in the 1970s and 1980s (cf. Beaugrande 1980-81).

So much for Widdowson's "right" methods: anodyne doses of "psycholinguistics" and "ethnographic inquiries" wherein the "text" is almost as narrowly conceptualised as it was in 1973. Issues like the politicisation of discourse on worker safety nowhere appear but in the Preface [10]. The book ends with the already noted patronising admonition scolding us "academics" to "conform to the conventions of rationality, logical consistency, empirical substantiation, and so on that define authority" (173) - of which the book itself hardly seems to shine as an epochal demonstration. We are despondently left in limbo with the echoes of the wholly unintentional irony in an earlier dismissal of Wodak and her study group (Titscher, Meyer, Wodak, and Vetter 2000):


[19] There seems little point in providing such a complex theoretical and procedural apparatus without demonstrating how it actually works. (144)


Res ipsa loquitur.




Beaugrande, R. de. 1980-81. Design criteria for process models of reading. Reading Research Quarterly 16, 261-315.

Beaugrande, R. de. 1980. Text, Discourse, and Process: Toward a Multidisciplinary Science of Texts. Norwood, N.J.: Ablex. 

Beaugrande, R. de. 1984. Text Production. Norwood, N.J.: Ablex. 

Beaugrande, R. de. 1991. Linguistic Theory: The Discourse of Fundamental Works. London: Longman. 

Beaugrande, R. de. 1998a. On "usefulness" and "validity" in the theory and practice of linguistics: A riposte to H.G. Widdowson. Functions of Language 5/1, 87-98.

Beaugrande , R. de. 1998b. Performative speech acts in linguistic theory: The rationality of Noam Chomsky. Journal of Pragmatics 29, 765-803.

Beaugrande, R. de. 2001. Interpreting the discourse of H.G. Widdowson: A corpus-based critical discourse analysis.. Applied Linguistics 22/1, 104-121.

Beaugrande, R. de, and W.U. Dressler 1981. Introduction to Text Linguistics. London: Longman.

Boston, T. 1998. Life, liberty and the pursuit of resourcism: An ideological analysis of the "Wise Use Movement". The Trumpeter (internet journal). 

Chomsky, N.S. 1957. Syntactic Structures . The Hague: Mouton.

Dijk, Teun van. 1972. Some Aspects of Text Grammars. The Hague: Mouton

Dijk, Teun van, Ihwe, Jens, Petöfi, János & Rieser, Hannes. 1972. Zur Bestimmung narrativer Strukturen auf der Grundlage von Textgrammatiken. Hamburg: Buske

Halliday, M.A.K. 1973. Explorations in the Function of Language. London: Arnold.

Halliday, M.A.K. 1994. Introduction to Functional Grammar. London: Arnold.

Harris, Z.S. 1951. Harris, Methods in Structural Linguistics. Chicago: University of Chicago Press.

Harris, Z.S. 1952. Discourse analysis. Language 28, 1-30 and 474-94.

Harris, Z.S. 1957. Co-occurrence and transformation in linguistic structure. Language 33/3, 283-340.

Rampton, S., and Stauber, J.. 2001. Trust Us, We're Experts. New York: Tarcher/Putnam.

Seidlhofer, B. (ed.) 2003. Controversies in Applied Linguistics. Oxford: Oxford University Press.

Titscher, S, Meyer, M., Wodak, R., and Vetter, E. 2000. Methods of Text and Discourse Analysis. London: Sage.

Widdowson, H.G. 1973. An Applied Linguistic Approach to Discourse Analysis. PhD thesis and "Unwritten Book".

Widdowson, H.G. 1992. Practical Stylistics. Oxford: Oxford University Press.

Widdowson, H.G. 2004. Text, Context, Pretext: Critical Issues in Discourse Analysis. Oxford: Blackwell.

More information about the Critics-l mailing list