Negatives in written text   251
16 Negatives in written text
Adriana Pagano
INTRODUCTION
Studies of negation have traditionally focused on morphological, syntactic and logical aspects, without considering use or meaning in context, indeed, thexe^haye been relatively few studies dealing with negatives from a >^\gragmatic perspective^ and still fewer attempting to systemize the uses of negaliolirAmohg'tne latter, Tottie (1982, 1987) has proposed a chissifi^ation_ _pf Jhe_uses of negatives in both oral and written language.
I collected examples of negatives from written texts and approached my data from three different perspectives, corresponding to the three language functions pointed out by Halliday (see Pagano 1991). From an interpersonal perspective, I analysed the role of negatives in the interaction between writer and reader in order to see why negatives appear in texts. From a textual perspective, I looked at the role of negatives in both the micro- and the macrostructure of the texts. That is, I analysed how negatives relate to adjoining clauses and to the text as a whole. Finally, from an ideational perspective, I compared overt (i.e. negatives having a formal marker of negation such as not, no, nowhere, etc.) and covert negatives (i.e. propositions expressing a negative meaning but having a positive form, such as / forgot), in order to see whether equivalent forms of overt and covert negatives (e.g. / did not remember -1 forgot) represented similar ways of expressing content in language. In the present chapter, I will concentrate on the first of these three perspectives: why do negatives appear in texts?
NEGATIVES FROM A PRAGMATIC PERSPECTIVE
Before discussing negatives in the interaction between writer and reader, I will first define the object of my study, which is implicit negatives or denials, as Tottie (1987) labels them.
What then are implicit negatives or denials? According to Tottie, negatives are used for two main purposes:
to reject suggestions (including refusals) to deny assertions
from which she draws two categories: 0
Rejections (including Refusals) Denials
The main difference between Rejections and Denials, Tottie claims, is that rejections express volition on the part of the sender, for example:
A: Would you care for a drink? B: No, thanks.
whereas denials do not; they are concerned with facts; they just state that an assertion is not true, for instance:
A: So you are still living out there.
B: No, I am not. I have rented a flat near the bank.
This explanation, however, is not satisfactory in its differentiation of the two negative uses, since it might be argued that volition, a quality associated exclusively with rejections by Tottie, may also underlie denials. That is, there is also volition in expressing a denial.
The difference between rejections and denials may be better explained from a func^pnal^perspecjtiye, considering the function which is predominant in each situation.
If we regard rejections and denials from the perspective of language function (Halliday 1970, 1973), we note that the language component that predominates in each of these uses is different. In denials, the ideational component is predominant: when we deny something, we are concerned with expressing our view on a particular fact, that is, with whether things are one c, yy. way or another, suchas: «J
A: Then finally she got what she wanted.
B: Well, I wouldn't say that. She never wanted to break with him. Things just happened that way.
B's denial of A's assumption (the woman in question wanted to end her relationship with her boyfriend) has a strong ideational component: B wants to correct A's view of a fact; the truth or correctness of the fact is more relevant than the interpersonal element in the conversation. However, if the conversation went like this:
A: So the party is at 9. Shall I bring something to eat or ... ? B: No, thanks. Don't worry. We'll have pizza.
In saying 'no', B is assuming a role in the conversation; he is providing an answer to A's offer (interpersonal function). There is certainly an ideational component (A does not need to bring anything to the party), but the jnteracjyLgnal function (a rejection) is the one that predominates. Thus, taking into account the predominant language component in a particular instance of language use, rather than the notion of volition, we can posit in agreement with Tottie that rejections and denials constitute two different categories of negative use.
252  Advances in written text analysis
What denials negate, Tottie further states, can be either an explicitly stated assertion, as in
There are two kinds of waste producers: those that produce inorganic waste and those that do not.
or an assertion that is somehow implicit in the contort of the interaction, generally an assumption by the producer of the denial with respect to his interlocutor's beliefs or expectations, for example:
The most significant departure in the CEELT examination is that video recordings are used in the Oral and Listening Comprehension section. These are clips from actual lessons and not specially scripted. For the Oral test the clip acts only as a stimulus to interactive communication in groups of three.
(MET, 15 March 1988: 43) In this example, the denial 'not specially scripted' refers to a proposition which has not been explicitly asserted, as would be the case, for instance, if the paragraph had read:
In previous tests, the clips were specially scripted by our language specialists; this time they are not.
where an assertion is first presented and then denied. In the example taken from MET, the denial refers to an expectation wMchn^Ke writer assumgsJhis reader has concerning the scripts of the tests.
At this point, a brief explanatory note concerning Tottie's terminology is necessary. Her terms explicit and implicit denials are perhaps a bit misleading in that what is explicit or implicit is not, as might be expected, the denials themselves, but the^jtfopositions that arej3ejng_denie^. In other words, the explicitness/implicitness criterion used in her classification refers to the thing that is denied and not to the negative itself. Thus, in an exchange like:
A: Has the garbage been emptied?
B: You know bloody well I've been out all day, how could I have emptied the garbage can?
B certainly produces what is commonly called an implicit negative answer, meaning 'No, I haven't done it'; however, it is not what Tottie (1982) considers as jmjrficitjiemgl. Indeed, it is not a denial at all in Tottie's terms, since there is no formal marker of negation, such as no, never, nothing, etc. An implicit denial, according to Tottie, is a denial of a proposition which has not been explicitly formulated in the text. For example:
WHY MOSQUITO REPELLENTS REPEL
They are not substances that a mosquito somehow finds distasteful. They jam the mosquito's sensors so that it is not able to follow the warm and moist air currents given off by a warm-blooded animal.
(Wright 1975: 104)
/
/    ^Negatives in written text
253
Here, the writer is denying what the reader might believe in relation to repellents: they are distasteful to the mosquito. This is an idea implicit in the text, inferable from the text. This is what Tottie calls jmjjjiziulenial.
Implicit denials then are denials which originate as a product of an assumption by the producer in relation to his/her interlocutor's beliefs. Being denials of implicit assertions, they reveal aspects of discourse which are not explicit- In other words, they reveal the process going on in the<producer's** mind when constructing a communicaTTve message.- ~
Among all the uses of negatives, implicit denials seem particularly interesting, as tney raise questions as to what the producer has in ftiind to cause him/ "her to produce a denial, why a particular assertion should be implicit in a particular situation, why a denial fits the context in which it appears.
In order to investigate these questions, I analysed examples of implicit denials taken from a wide range of written texts: text-books, magazines, journals, scientific papers and newspapers. The criterion used to select implicit denials in the written texts is in line with that adopted by Tottie, that is, negatives in which the proposition denied did not appear in the text. Since in this study I have concentrated on denials, the data chosen from the source texts are those falling within what Quirk et al. (1985$ rail 'clause negation', namely instances in which the whole clause is negated, either through the main verb, for example he doesn't know any Russian, or through means other than verb negation, such as he knows no Russian.
WHY DO DENIALS APPEAR IN TEXTS?
I have repeatedly mentioned the term 'interaction' between writer and reader. But what are the characteristics of this interaction? Unlike in spoken discourse, in written text there is no physical receiver of the producer's message at the moment of composition. Nevertheless, the writer replaces the absence of a physical interlocutor by a mental representation of the reader. That is, the writer creates a picture of the reader, who thus becomes an 'ideal ~ puMUJ^ts reader', and attributes to this reader certain experience, knowledge, opinions and beliefs on the basis of which the writer.builds his/her message. (See Coulthard, this volume, pp. 4-5..)„„^e4» e i**
<As^Widdowson (1979)^states, in writing texts 'the writer assumes the roles of both addressor and addressee' and thus, as s/he writes, the writer thinks of the reader's possible reactions, anticipates them and acts accordingly. Whenever the writer feels the text may raise a doubt or leave a question unanswered, s/he tries to provide the information s/he thinks the reader is expecting. As the writer somehow assumes what the reader's questions and f , expectations are, s/he tries to provide information about these.   ^9   ,# r ^«>*s
Therefore, in cases where certaiixjnformation is non-existent, the writer r "'"t<qBrfi-. can report that by means of denials of what was expected, for example:
In Trankle & Markosian (1985), Expert System Adaptive Control (ESAC) is described. The system consists of a self-tuning regulator augmented with
254   Advances in written text analysis
three different expert system modules: the system identifier, the control system designer and the control implementation supervisor. A real time version of the system has not been implemented.
(Automatica, 26 June 1989: 815)
In this piece of text, the writer talks about an expert system which has been developed by some researchers. The writer probably thought that the reader might be interested in knowing whether the system has been implemented.' That is probably why he tries to satisfy<the reader's question"; with the information expressed through the denial.
A writer is often aware that the message may contain parts which are dubious or likely to be misinterpreted. In order to be clear and not to mislead the reader, s/he will point to the ambiguous stretches and caned the! potentially wrong interpretations. This can also be done by means of denials, such as:
Anyone with a passion for hanging labels on people or things should have little difficulty in recognizing that an apt tag for our time is the Unkempt Generation. / am not referring solely to college kids. The sloppiness virus has spread to all sectors of American society. People go to all sorts of trouble and expense to look uncombed, unshaved, unpressed.
(TIME, 2 April 1990: 46)
Here, the writer denies what readers might have thought in connection with the expression 'unkempt generation', namely that he was referring to adolescents, an idea typically associated with the above-quoted expression.
Of course, for writers to deny certain ideas or expectations which they assume on the part of their readers, the propositions denied have to be somehow plausible or expected in the context of the text. As Miller and Johnson-Laird (1976: 262) state:
people do not ordinarily go about uttering such denials as 'George Washington is not a table' or 'Sealing wax is not a dog', even though they are perfectly true... . These denials seldom occur because their corres^ ponding affirmations seldom occur.   {j&j&t      C ■7$?e&to ~ CJcf*£*\4
That is to say, for a writer to deny a belief or an expectation, s/he has to have some reason to think that the reader(s) may hold that belief or expectation, as is the case of the idea of youngsters in connection with 'unkempt generation'.
EXISTENTIAL PARADIGM
According to Miller and Johnson-Laird's observation, then, a writer cannot deny just anything, but only assumptions which are plausible or acceptable in the context of interaction. Borrowing Brazil's (1985) terminology, we could say that all those assumptions that are plausible or probable in a particular context of interaction constitute an existential paradigm. Why paradigm? Paradigm implies a group or set of linguistic items that are related in such a way that they may substitute each other in a given context. What determines
f *
Negatives in written text 255
the inclusion of the items within a single set is a series of conditions that all the members have to fulfil. Existential implies that what determines the inclusion of items within a set is experiential factors such as shared experience or understanding in relation to a given environment.
Brazil establishes a difference between the general and existential paradigms. The former consists of the conditions in the language system that limit the number of elements that can fill a slot. For instance in
A: Which card did you play? B: The Queen of Hearts.
in B's response, the slots filled by 'Queen' and 'Hearts' can be linguistically filled by many possibilities in the general paradigm of the language, for example the Fourteen of Lions. But, for these choices to be meaningful in this context, they have to be part of the existential paradigm, that is they have to be regarded as possibilities actually available in the given situation, which, in this particular case, they are not. Here, the limitation to the choices for these two slots is extralinguistic, imposed by the fact that in the real world, a pack of cards has four suits (none of them being 'Lions'), each of which has a determinate sequence of cards (in which 'fourteen' is non-existent).
If we take these two concepts (general and existential paradigm) and apply them to our study of denials, we can see that in a given situation, there are many linguistically possible denials that can be produced (certainly by linguists) and which make up the general paradigm, such as colourless green ideas don't sleep furiously, but as Miller and Johnson-Laird state, those denials are nonsensical, because those propositions could never take place in a real-world context. Thus, Jhejsetof propositions that could be denied in a given context is limited by the propositions which are experientially possible . m^thaTjcontext (existential paradigm). We could then define existentfob, paradigm as a set of assumptions which are experientially linked in a certain context. In order to clarify this, let us go back to the example on page 254, in which the writer writes about the 'unkempt generation'.
This expression is, in our culture, associated with ideas such as hippies and rebellious, long-haired adolescents. These ideas are part of an experience in our culture. If we hear or read 'unkempt generation', we expect the speaker/ writer to be referring to youngsters. Thus, we could say that the existential paradigm of the idea of 'unkempt generation', that is, the assumptions experientially linked with that idea, are matters related to young people, adolescents. Therefore, the idea of referring to college kids can be plausibly denied in the context of 'unkempt generation', since 'college kids' is paifcof the existential paradigm of this context. Had the writer written / am not referring solely to politicians we, as readers, would have found this idea odd or incomprehensible, certainly unexpected, which would have made us read on to find out the link between 'unkempt generation' and 'politicians'. The denial would not be aplausibleprie in this context. For us, 'politicians' is not -r~\j- within the existential paradigm of 'unkempt generation'.
I
pot
256   Advances in written text analysis
It is interesting to notice that when a denial is expressed, the producer is projecting a world in which what is denied is accepted, that is, in which there * is an understanding that the producer and his/her readers accept the proposition being denied. For instance, when somebody, commenting upon a wedding, • says The bride was not wearing a white dress, that person is projecting a world in which brides normally wear a white dress in contrast to the one he is talking about. Similarly, when he says The bride was not wearing a blue dress, he projects a world in which there was a reason to expect that she would wear blue, either because it is a custom in the group or because there was reason to believe that this particular bride would wear one. This has significant implications which are generally expjoitedjn certain mass,media. For example, if a bottle of juice is advertised as having no sugar, it is because producers assume people believe juices typically contain added sugar. If the label on the bottle read no fish-bone, people's first reaction would be one of surprise, because in our society nobody expects a juice to contain fish-bone. But then the very denial would project a world in which at least some other juices contain fishbone. This can actually create an expectation in the consumer's mind, as they may start to wonder whether the juice they habitually buy contains fish-bone, a substance certainly undesirable and perhaps prejudicial to health, since nobody would advertise something as not having a healthy or positive thing.
On some occasions, the writer provides a denial or idea s/he wants to correct after an assertion stating what s/he considers to be the right choice. For example:
Menstrual changes were significantly related to the intensity, not the activity, of running.
(Runner's World, June 1985: 29)
Making friends with the neighbouring Indians, he argued that the land belonged to them and not to the king or to the Massachusetts Bay Company. The Massachusetts government decided to deport him as a dangerous character.
(Current and Goodwin 1975: 12)
In these cases the writers are jnaking explicit the choices or jej;jsjential paradigm from which they selected the option asserted. Insojtoing, they are Y/ making sure the reader learns about the other members of the existential paradigm which they consider wrong. For instance, in the example about the Indians, the set of possible owners of the land (that is, the existential paradigm in this context) includes the king, the Massachusetts government and the Indians. On stating that the land did not belong to the first two, the writers are adding after their assertion a piece of information to make sure these two options are known to the reader. If the writers did not do this, the reader would probably never learn about the discarded options. And, usually, these options are added to the text because they are significant to the discussion of the topic.
T
Negatives in written text 257 Compare these two alternatives of a magazine advertisement. PCB CAD/CAE
TO MAKE YOU MONEY.
PCB CAD/CAE
TO MAKE YOU MONEY . . . NOT TAKE YOUR MONEY.
(Computer Design, July 1983)
In this case, the second version (the authentic one) makes explicit the option or choice that is being rejected. This is important, because the option denied represents a disadvantage with which the advertisement's offer is contrasted."
SCHEMATA CC
cr" (
v&vJffi ©. rm At «*'<v^
Besides discussing the set of choices a speaker has in a given context in the form of paradigms (existential and general), we can also regard this set of choices in terms of schemata.
Schemata or schemas are 'packets of information stored in memory representing general knowledge about objects, situations, events or actions' (Cohen et al, 1986: 27).
Why are schemas relevant to our discussion of denials? If, as has already been said, we can deny propositions which ar&^gusjple^in a given context and if schemas determine our expectations in connection with a given situation, then when referring to a particular schema, we can only identify propositions which are assumed to be part of that schema. Let me clarify this with an example: a picnic schema."
If somebody comments after a picnic The picnic was nice but nobody took any food, it is because they expected the people going picnicking to take food. 'Food' is a defining element in the schema of a picnic. If, on the other hand, somebody commented The picnic was nice but nobody watered the grass, this would sound an odd comment on a picnic: when you go to a picnic, you do not normally water the grass of the place where you stay. This seems to indicate that the things we can plausibly deny concerning a schema have to be considered as likely to be part of that schema.
We should always bear in mind that even the more general schemas are culture-specific, which implies that the values which a cultire considers part of a schema may be different from the ones considered by another culture. Let us take the wedding schema. For us, a traditional wedding involves the bride and bridegroom, a ceremony performer, etc. (fixed values) and also a white dress for the bride, throwing confetti, music, etc. (conventionally optional ones). So, if somebody remarked about a wedding:
the bride wasn't wearing a white dress
this denial would represent the expectations the producer had in connection with the bride. If, however, that person remarked
258   Advances in written text analysis
the bride didn't turn over the cash gifts to her parents
this denial would sound odd to us, because giving the cash gifts to the bride's parents is not part of our schema of a wedding. But for a Korean, this denial is meaningful, because that practice is customary in his culture and constitutes an element in the schema of a wedding.
All this indicates that for people to deny something, they assume that they and their interlocutors share a common world in which certain beliefs and expectations are usual. Taking up the concept of ideal>eader, all this implies that the writer attributes to his/her ideal reader certain knowledge (schemas) and beliefs or ideas specific to the topic being dealt with. Taking those attributes for granted, the writer can build a message aimed at a target reader.
DATA ANALYSIS
In order to see how denials appear in this process of text construction, I selected and considered a reasonable number of denials and tried to see which ideas or propositions were denied and whether there was any reason for the writer to deny them. Remember that implicit denials, as their name states, refer to propositions which are not explicit in the text. The fact that they make no reference to an explicit proposition, however, does not mean that they appear out of the blue, without any connection at all to the topic being developed. They occur because there must be some reason why the writer feels the need to use a negative.
From the analysis of the data, I extracted four reasons why the writer does this and I classified them into four categories:
(1) Denials of background information: that is, denials used when the writer assumes that the reader entertains certain mistaken ideas from his previous background knowledge.
(2) Denials of text-processed information: that is, denials used when the writer assumes that the reader could derive a wrong idea from the text. ^
(a) Denials used to prevent an erroneous inference from,text to come.
(b) Denials used to correct an idea already processed/in the text.
(3) Unfulfilled expectations: that is, denials used when the writer wants to express an unfulfilled expectation of which s/he makes the reader co-participant.
(4) Contrasts: that is, denials used to compare or contrast two or more items.
In order to clarify these categories, let us consider each of them with examples from my data.
Denials of background information
In this class, the propositions denied by the writer are ideas which s/he assumes the interlocutor may, irrespective or independent of the text itself,
Negatives in written text 259
entertain in connection with some aspect of the topic being dealt with, for instance:
Another important point to remember is that sexual orientation is not completely permanent. Especially in adolescence, but also throughout adulthood, sexual orientation can change.
(Coleman 1981: 217)
The idea that sexual orientation is permanent is attributed to the reader. The writer, being familiar with the topic and with ideas generally held about it, such as the one denied in the example, feels the need,to correct those wrong views through a denial and present his own view afterwards. Similarly in
Oat-bran muffins alone aren't going to save you. Eat a high-fiber diet with a variety of foods. Emphasize vegetables, fruit and whole grains.
{Runner's World, March 1990: 68)
The writer denies a view which gained popularity through mass-media publicity: the miraculous power of oat-bran to reduce cholesterol, and which he knows his audience is likely to entertain. As the wrong idea (from the writer's viewpoint) is a popular and widespread one, he assumes his reader believes or at least is familiar with it. Thus, he must explicitly correct it.
The idea denied can also be a specific one (e.g. in academic writing), which only those who are familiar with the basic assumptions and theories of the ^sgeclfic area xan actually have, as they belong to the community which sustains those ideas. Hence in:
Design is not art. It is also not engineering, and it is not science. It is time to recognize this and distinguish the differences. Design is not separative, it is integrative. One of the hallmarks of design is its penchant for integration.
(Owen 1989: 4)
a number of misconceptions about design, which the writer knows his reader may entertain, are denied. For someone outside the field of design, the denials and/or the necessity for them may sometimes not be fully understood.        ^ '
This category of denials is the commonest in the data. Here are some more J examples.' a^^j/k-t*,**^
A text is a semantic unit, not a grammatical one. But meanings are realized through wordings ...
(Halliday 1985: xvii)
Also to learn. Chamorro owes her election not to any natural gift for leadership but to her married name. Though graced with regal poise and an engaging personality, she has had little experience in public life.
(TIME, 12 March 1990: 13)
One reason is to help you learn new, permanent eating patterns. While restricting your calories certainly will help you lose body fat, weight loss
f -
/ / • r roptt
260  Advances in written text analysis
and dieting cannot go on forever. They're merely temporary therapeutic measures to help you attain a desirable body composition.
(Runner's World, January 1987: 36)
It must be remembered that, as I have already stated, what is denied must be within the range of possibilities that can be plausibly denied (existential paradigm). Thus in / '
He was carrying the .25 when the cops arrested him on the street the following day. He wasn't wearing colors; few members do any more, since gang emblems are as open an invitation to arrest as carrying a semiautomatic rifle. But just the fact that he was dressed in low-slung black trousers, Nikes and Pendleton shirt gave him away.
(TIME, 18 June 1990: 22)
the statement that the boy was not wearing colours is denying the idea that is generally assumed of a gang: they usually have special or distinctive clothes. In fact, the writer goes on to explain that wearing colours is a practice no longer followed by gang members. Wearing colours is an idea that is plausibly related to gangs. Had the writer said he wasn't eating peanuts, it would certainly have been unexpected, unless the writer later explained that eating peanuts is a usual practice among gangs, that is, unless it were part of the existential paradigm of gangs. V'^,„, „ _ rJi- f>( at-**
Denials of text-processed information
This class of denials involves propositions (i.e. the ideas or beliefs) which tne writer thinks the reader could wrongly infer from the text. This implies an awareness by the producer of which points in the text will be dubious or ambiguous for the reader. Therefore, out of solidarity with the reader and because the writer is interested in the reader's understanding and eventually supporting his/her view, the writer detects those misleading parts in the text and through denials cancels the wrong inferences, for example:
If a male increases his chances of reproduction through this type of cooperation because the favor is returned later on, the behavior is referred to as reciprocal altruism. In many species reciprocal altruism appears to have evolved in response to situations where it is difficult, if not impossible, for a solitary male to successfully mate with a female. This interpretation is not universally accepted. Scott Kraus and John Prescott ... suggest... that the males are not cooperating but rather are competing with one another for access to the female.
(Wursig 1988: 81)
In this example, the writer presents a theory accounting for the sexual behaviour of whales. The way in which Wursig presents the theory may lead the reader at that point in the text to think the theory is an accepted or probable one. As this is not the case, the writer points out the relativity of the
Negatives in written text 261
view presented, denying what he thinks may be the reader's possible wrong inference: that the theory is generally accepted. Within this kind of denial, we can distinguish between (a) and (b):
(a) Denials in which the ideas denied may actually be ideas or reactions which the writer anticipates or expects the reader to have in connection with what
s/he is going to say in the text, for instance:      -..      , A ■
__? , i ' " c
7 don't want to sound sentimental but they've been saving up to come here to see me and they'd see me talk the songs ...'
(The Sunday Times, 22 April 1990: 1)
Here the speaker cited in the text anticipates the reaction that his words may produce in his interlocutor (the reporter that interviewed him); he denies what he believes the listener will think about him.
(b) Denials in which the proposition denied has been suggested by previous parts of the same text as in:
'Relatively few people have applied for studies in the exercise line', says Wood. 'The number's not negligible, but proportionately it's rather small. Drugs are ridiculously over-represented, and that has to reflect the importance of the drug companies . . .'
(Runner's World, May 1990: 1)
It has been estimated that nearly 50% of recent marriage cohorts will experience marital dissolution .... However, these figures do not necessarily mean a massive rejection of marriage and family life, as most individuals who experience marital dissolution eventually remarry.
%\/c.u^+AA--- (Teachman and Heckert, 1985: 185)
In these examples, there is some idea in the text which, the writer thinks, is likely to lead the reader to a wrong interpretation. That is why that idea has to be expanded or clarified; for example, if the writer states that graffiti became more elaborate and less vulgar, the reader may very well come to think that from that moment onwards all graffiti were elaborate and critical, which is not true. Hence the writer's cjarificatory denial.
Unfulfilled expectations
Implicit denials can also represent unfulfilled expectations on the part of the writer and the reader, as the former assumes his/her expectations are also the latter's. Let us consider the following example:
To control the development of a plant, light must have some effect on the developmental blueprints, the genes. Yet whereas the biochemical reactions involved in photosynthesis have been described at length, it is not at all
-k
í
■
262  Advances in written text analysis ■ 1
clear how light influences the expression of plant genes. Six years ago we set out to clarify the biochemical basis for photomorphogenesis by beginning with the light-responsive genes themselves.
(Moses and Chua 1988: 64)
The writers here acknowledge some information (the biochemical reactions in photosynthesis have been described at length) but then indicate that some information which for them is relevant and, they assume, for their reader too, is not clear. We could say that the writers have an expectation connected with something that is significant for the field of research and believe that their reader, being someone interested in that field of research, has the same expectation; therefore, they inform the reader about what they expected.
It could be claimed that some expectations seem to be the writer's, in which case the denials give the reader access to the writer's own thought processes. However, we must always bear in mind that the writer writes for an ideal reader and the very fact that s/he includes an expression of an unfulfilled expectation in the text may indicate that s/he thinks that information about that unfulfilled expectation (be this missing information or an excuse for not presenting or dealing with some aspect of the topic in question) is relevant for the ideal reader.
Similarly, in y cmi^'^
■ • Bowheads are also believed to feed at the bottom (usually at depths of less than 60 meters), but, it is not clear how, equipped with long and finely fringed baleen, they are able to do so. We have seen bowheads surfacing with muddy water streaming from the sides of the mouth, a behavior that in gray whales is clearly associated with bottom feeding.
(Wursig 1988: 79)
But Bart doesn't illuminate why Kerkorian does what he does. He offers a few new anecdotes about Kerkorian's gambling habits, but the financier never comes alive. Bart seems content with his conviction that the MGM boss lives simply to be a trader and is not a manager.
{Business Week International, 23 July 1990: 8)
Generally, an unfulfilled expectation leaves implicit the idea that what is missing, absent, etc., should be otherwise.
Sometimes, the writer tries to respond to.expectegjgns wJlufiLarejinly the reader's expectations. That is, the writer did not expect that but thinks the reader probably will. This is clearly seen in prefaces, introductory or aclaratory notes, in which the writer states his/her intention concerning the text and the scope of his/her work. Consider these examples:
This article attempts not grand solutions but rather a clarification of some of the theoretical differences between two major camps in the current debates, recognizing that in these debates political commitments often precede and determine theoretical positions.
(Valverde 1989: 237)
Negatives in written text 263
The purpose of this short section... is to explore a few points prompted by material in the main part of each chapter . . . No attempt is made to be comprehensive and some comments are concerned with areas marginal to the main themes.
(Pears 1985: 9)
Both writers exempt themselves from presenting information which their readers could otherwise have reasonably expected.
Contrasts
To this class belong those denials which appear as implicit contrasts between two or more things, for example:
For past generations, lifestyle was the leading pharmacopeia. They had no antibiotics, no cures for infectious disease. They had to rely on their manner of living to preserve their health.
(Runner's World, February 1990: 16)
Here, there is an implicit comparison between the past and the present, and the differences are pointed out. These are actually the denials indicating the absence of antibiotics and cures, which we now have. In
About three fourths of the U.S. population is concentrated in urban areas. People growing up in the inner city often have little exposure to or opportunity for leisure/recreation experiences in a natural environment. Adult urbanites, however, are more likely than rural residents to be recreation/leisure users of public outdoor areas. Yet, they were not provided the opportunities during youth to learn the skills and knowledge which would allow them to understand the dynamics of the natural environment.
(Journal of Physical Education, Recreation and Dance, April 1990: 49)
the comparison takes place between the people living in town and the people living in rural areas. In
At Sudbury Valley there is no curriculum. There are no academic requirements. There is no evaluation of students except when requested, no grades or other devices to rank them from best to worst. There is no school-imposed segregation of any kind - not by age, not by sex, not by ability. Students are free to move about at will, using the school's laboratories, workshops, library, playground, and other resources.
(Phi Delta Kappa, May 1984: 609)
the contrast is established between, traditionahschools and Sudbury school,
which claims to be special.
264   Advances in written text analysis Another category?
In a small number of cases in the data, denials apparently have a different purpose from those already analysed. These specific denials contain a modal verb meaning possibility and are followed by a restriction which actually opposes them, for instance:
Saabs may not look large. Yet the Saab 9000 is the only imported car in the r USA rated 'large' by the Environmental Protection Agency. > ti I'■'<;J\..
(Business Week International, 12 March 1990: 1) 8
Through the denial the writer is actually admitting, or better, conceding the ♦ fact that Saabs do not look large. It is as though he were saying: 'O.K. I agree. Saabs may not look large. But...' And there comes a restriction implying some Saabs are large. Another example is:
Solar technology may never eclipse conventional power sources. But it already promises the children of Africa a brighter future.
(Business Week International, 23 July 1990: 15)
The same process takes place here. The writer admits something but then presents an alternative which reduces the effect of the denial.
As the number of examples having similar features to these two examples above is very small, we cannot at the moment formulate or state any features of this category but just point to its potential existence.
I
CONCLUSIONS
In analysing the claim of the writer when producing a denial, we have so far dealt only with the interpersonal perspective of denials, that is, denials in the interaction between writer and reader. We can also approach negatives from a pragmatic perspective and see how denials relate with adjoining clauses. For instance, we can analyse the environments or clause relationships in which denials appear most frequently, as in Denial-Correction,
What Lithuania is experiencing, therefore, is not betrayal, nor is it appeasement. It is tragedy.
(TIME, 16 April 1990: 52)
and see whether the accompanying member of the denial is actually predicted or anticipated by the denial. That is, whether after the denial in the Denial-^J^rrec^nptttterii, we, as readers, expect a correction.
From an ideational perspective, we can also ask ourselves whether apparently equivalent forms such as
These two insects do not belong to the same species. These two insects belong to different species.
I
Negatives in written text 265
represent the same way of expressing content in language and for what purposes would one be used instead of the other.
A final goal could be to integrate the three perspectives and see the three language components at work in negatives.
APPENDIX: TEXTS USED OR REFERRED TO Journals
Automatica, 26 June 1989: 815.
Business Week International, 12 March 1990: 1; 23 July 1990: 8, 15. Computer Design, July 1983.
Journal of Physical Education, Recreation and Dance, April 1990: 49.
MET, 15 March 1988: 43.
Phi Delta Kappa, May 1984: 609.
Runner's World, June 1985: 29; February 1990: 16; March 1990:.68; May 1990: 1. The Sunday Times, 22 April 1990: 1. TIME, 2 April 1990: 46; 16 April 1990: 52.
Articles and books
Brazil, D. (1985) The Communicative Value of Intonation in English, Discourse Monographs
No. 8, Birmingham: University of Birmingham. Cohen, G. et al. (1986) Memory: A Cognitive Approach, Philadelphia: Open University Press. Coleman, E. (1981) 'Counselling adolescent males', The Personnel and Guidance Journal
(December): 215-18.
Current, R. and G. Goodwin (1975) A History of the United States to 1877, New York: Alfred Knopf.
Halliday, M. A. K. (1970) 'Language structure and language function', in J. Lyons (ed.), New
Horizons in Linguistics, Harmondsworth: Penguin Books, 140-65.
- (1983) Explorations in the Functions of Language, London: Edward Arnold.
- (1985) An Introduction to Functional Grammar, London: Edward Arnold.
Miller, G. and P. Johnson-Laird (1976) Language and Perception, Cambridge: Cambridge
University Press. ° ' ~*j
Moses, P. H. and N. Chua (1988) 'Light switches for plant genes', Scientific American
258/4 (April): 64-9.
Owen, C. H. (1989) 'Design education in the information age', a speech given at Korea Institute
of Technology's Exposition and Design, 20 November, Seoul. Pagano, A. (1991) 'A pragmatic study of negatives in written text', unpublished MA diss.,
Florian6polis: Universidade Ferderal de Santa Catarina. Pears, N. (1985) Basic Biogeography, London: Longman.
Quirk, R. et al. (1985) A Comprehensive Grammar of the English Language, London, Longman.
Teachman, J. and A. Heckert (1985) 'The impact of age and children on remarriage',
Journal of Family Issues 6/2 (June): 185-203. Tottie, G. (1982) 'Where do negatives come from?', Studio Linguistica 36/1: 88-105. - (1987) 'Rejections, denials and explanatory statements - a reply to Fretheim', Studia
Linguistica 41/2: 154-63. Valverde, M. (1989) 'Beyond gender dangers and private pleasures: theory and ethics in the sex
debates', Feminist Studies 15/2: 237-54. Widdowson, H. G. (1979) Explorations in Applied Linguistics, Oxford: Oxford University Press. Wright, R. (1975) 'Why mosquito repellents repel', Scientific American 233/1 (July): 104-11. Wursig, B. (1988) "The behavior of Baleen whales', Scientific American 258/4 (April) 78-85.
t-LL