Negative evaluators

Negative evaluators serve to place the narrative action against a background of events that did not, in fact, happen, but that could have happened. Examples of this are given in the texts analyzed below. One effect of placing narrative events against a wider contextual background is to heighten the dramatic impact of a narrative – by making the reader aware, for example, of how lucky or improbable the actual sequence of events was.

The quantitative data reveal that use of negatives is considerably higher in the Times articles than in the China Daily articles. There are 107 instances of negative evaluation in The Times texts, compared to just 42 in the China Daily. Negative evaluation occurs on average once every 187 words in The Times, but just once every 395 words in the China Daily.

As we have seen, these differences are statistically significant. A closer look at some of the texts displaying a very high or very low incidence of negative evaluation gives insight into why they may exist. The Times text showing the greatest degree of negative evaluation (nine instances altogether, excluding the headline) is Times article 38. Headlined “School will not punish Straw's son,” it is an account of how the headmaster of the school at which Home Secretary Jack Straw's son William was a pupil said he would not punish the boy for selling marijuana because he had been punished enough already by the media publicity. Interestingly, there is even a negative in the headline, which gives some indication of the way in which things will go.

The negative evaluators found in the text are set out in Table 6.

The first thing that is striking about this article is that the whole narrative is predicated on a negative. The very first paragraph says that the headmaster has no plans to suspend or discipline the teenager. Selling marijuana is not very serious, it is not the end of the world, and while Jack Straw wishes the incident had not happened, police are not taking any action, and anyway, say William's fellow pupils, the boy doesn't deal and would never buy drugs. The entire story is in fact about nothing happening. But if there is nothing happening, where is the story? The story is in the denial; which is to say, it is in the negative evaluation. By beginning the article with the statement that the headmaster has no plans to suspend or discipline the teenager, the journalist immediately raises in the reader's mind the possibility that the headmaster was seriously considering doing so but has decided not to. In other words, the story is in what might have happened but didn't. The reader is left wondering what would have happened if the headmaster had decided otherwise – and is also left with the feeling that because the head teacher has spoken out to say he was not intending disciplinary action, and because a newspaper thought that decision worth reporting, the possibility that the headmaster might have decided otherwise was a very real one.

Much the same is true of the other, subsidiary negatives that support the remainder of the story. When Jack Straw says he wishes the incident had not happened, it reminds readers that it did and emphasizes the fact that it was a serious matter. When a pupil insists there is no stigma about buying drugs, it makes readers feel that there is, or why does the pupil even need to bother denying it? When the police are said to have recommended no action, it makes the readers realize that they could easily have done so, and that this would have been a serious matter.

In all these cases, negative evaluation is being used to contrast what is with what might have been. But this contrast is being achieved in different ways. In some instances, it is used to create a reportable narrative description of a situation in which there is no real narrative action at all. Thus, in the clause The headmaster at William Straw's school says he has no plans to suspend or discipline the teenager, the fact that the head teacher has no such plans itself becomes a positive, reportable act, and the fact that he is not taking punitive action is contrasted with the possibility that he might have decided otherwise. In other instances, negative evaluation is used to turn a denial (rather than an absence of action) into a reportable act, thereby making it possible to state precisely that which is being denied. Thus, a schoolmate of William Straw says, Will doesn't deal. He would never buy drugs – and by reporting those words, the journalist raises in the mind of the reader the possibility that perhaps William might after all. In yet other instances, negative evaluation is being used for dramatic emphasis. I really don't think it is the end of the world, says the head teacher, and by saying so somehow manages to imply that it almost was. It was inevitable for a parent to … . wish the incident had not happened, Jack Straw is indirectly reported as saying, emphasizing once more how serious was the fact that it did.

The reason this story is newsworthy, even though it is actually about nothing happening, is that the main protagonist is the son of an important and powerful man. The importance of the protagonist means the journalist considers it worthwhile to use certain linguistic techniques (including negative evaluation) to construct a reportable narrative even when very little has actually happened. The technique employed here of using a negative (or series of negatives) to do this is one that, I would argue, is commonly used by Western journalists. If there is a rumor about a scandal involving a famous person and it can't be proved, get the person in question to deny it, and you've got your story.

Is the pattern of use of negative evaluation found in Times article 38 repeated in other Times articles showing comparatively high use of evaluation? Essentially yes. Times article 5, which has seven instances of negative evaluation, is headlined “Blair condemns Diana stories.” A look at a few instances of negative evaluation in the article reveal that, again, it derives much of its drama and richness from the contrast between what is and what might have been. Thus: Blair's staff said his remarks should not be seen as criticism of Fayed constitutes a denial which itself becomes a reportable act and allows what is being denied to be repeated, prompting instant speculation that what is being denied might actually be the case. [Fayed] told a newspaper he … did not believe the crash was an accident emphasizes that there was more to the crash than meets the eye, by contrasting the current belief that it was an accident with the possibility that it was not. He [Blair] is not singling out any individual or enterprise, insisted one spokesman is another denial, leaving the lingering suspicion that Blair may be doing precisely what the spokesman says he is not doing, a suspicion reinforced by use of the verb insisted, with its rather negative overtones. Blair has not discussed his views of the speculation surrounding Diana's death with the Prince of Wales makes a “positive out of a negative,” like William Straw's headmaster is not going to take disciplinary action; the journalist is able to make something positively reportable out of the fact that Blair has not done something simply by reporting that he hasn't done it.

Much the same pattern is seen in Times article 40, headlined “Queen pops into haunted local,” which also contains seven instances of negative evaluation. Here, we are told that there was no sign of Nancy the resident ghost – a sentence that neatly establishes the apparent fact of the ghost's existence by saying she wasn't there. Later, the pub's landlady is quoted as saying, “I did not offer her a drink and she [the Queen] did not ask for one;” by saying so, she succeeds in efficiently making a narrative out of a complete absence of event by contrasting actual with potential happenings.

Although the Times articles generally use more evaluation than the China Daily texts analyzed, there are Times texts that display little or no evaluation. One such is Times article 48. The article, headlined “Relatives see film of sunken trawler,” is an account of an inquiry into the sinking of a ship, the Westhaven. The events being investigated are actually very dramatic, but the report is distant and impersonal: a mere recital of facts rather than a colorful narrative. Thus, The boat was dragged to the seabed so quickly that one of the liferafts was caught in the mast and a second failed to inflate. Mr Pattison and his crew, Alan Cunningham, 28, Chris Prouse, 23, and Mark Hannah, 30, died. There is no attempt here to get into the minds of the crew as they died; no “they could not get to safety because the liferafts failed.” The story is allowed to tell itself, and the suffering of the men is left to the reader's imagination. The contrast with the evaluatively rich articles analyzed above – which are actually reports of far less dramatic events – is great.

Negative evaluation, clearly, is a rich and effective linguistic device for heightening and enriching the drama of a narrative – and even for constructing a narrative out of little or nothing. It does this essentially by contrasting what is with what is not but might have been. There are a number of ways in which this is achieved in the articles so far considered: by making a positive out of a negative and creating a reportable narrative description of a situation in which there is no real narrative action at all; by turning a denial into a reportable act, thereby making it possible to state precisely that which is being denied; by reminding us of what might have been and so increasing dramatic emphasis; and by heightening dramatic tension by contrasting two different possible realities (as in Fayed's saying he did not believe the crash was an accident).

Negative evaluation is comparatively rare in the China Daily texts analyzed, and, as the quantitative data reveal, its distribution through the corpus of texts is comparatively clumpy. Of the 50 texts analyzed, 32 contain no negative evaluators at all, while the top five texts in terms of frequency of negative evaluation contain 52% of all instances of negative evaluation found in the corpus.

The China Daily text that displays the greatest use of negative evaluation is text 20. This text, headlined “Beijing tightens control over fireworks in city proper,” contains no fewer than eight instances of negative evaluation – almost one-fifth of the total for all 50 China Daily texts studied. The instances of negative evaluation that occur are set out in Table 7.

Clearly, there is an element here of “what is” being contrasted with “what is not, but might have been”: The officially arranged fireworks displays ought to have been a pacifier (what is not) but actually became a fuse sparking further trouble (what is); no fireworks-caused fires have been reported (what is), but they might well have been (what is not, but could have been).

Equally clearly, though, there are no examples in this text of negative evaluators being used to bolster or brighten a narrative, as is often the case in the Times articles – no turning of denials into reportable fact, no making a positive out of a negative to create a narrative where really there is none (unless the people's congress ruled that no firecrackers would be allowed is counted). The tone imparted by the negatives here is one of disapproval – almost as though the authorities, through their mouthpiece the newspaper, are saying, “This is what we warned would happen if you behaved like this.”

The China Daily text displaying the next highest use of negative evaluation is text 42, headlined “Mainland to export live chickens to Hong Kong.” The article is an account of attempts by the Chinese government to resume export of chickens from southern China to Hong Kong following a scare (prophetic, in the light of SARS) caused by an outbreak of avian flu. It contains five instances of negative evaluation: the chickens will not reach markets until Sunday; the price for live chickens is not expected to increase greatly; no case [of avian flu] had been detected either in humans or among poultry in Guangdong Province; no case of humans infected by the bird flu has been reported; although no cases of A H5N1 bird flu in human beings and chickens have been reported…. Guangdong will further tighten its testing.

What is notable about these cases of negative evaluation is that they are all, without exception, flat statements of fact, presented with the ring of absolute authority. China Daily text 42 in fact reads, to Western eyes, more like a government press release than an independent news report. The litany of no cases somehow has the effect of bludgeoning the reader into acceptance of what is said rather than, as is the case in Times articles rich in negative evaluation, raising in the reader's mind the possibility that what is denied may actually be the case.

The difference may have something to do with the way in which negative evaluation in the Times texts tends to be personalized, whereas in the China Daily it is not. In Times text 38, about Jack Straw's son, all but one of the instances of negative evaluation occur in the context of either direct or reported speech, and hence the denial which it represents can be attributed to an identified individual (the headmaster, Philip Barnard; Jack Straw himself; three unnamed but distinct fellow-pupils of William Straw) (see Table 6). Much the same is true of Times text 5, “Blair condemns Diana stories,” where, variously, Tony Blair's staff, Blair's spokesman (and indirectly through him, Blair himself), and Mohamed al-Fayed are all associated with denials.

In the few China Daily texts that display relatively high use of negative evaluation, this is not the case. In China Daily article 20 about firework controls in Beijing, negative evaluation is associated with, in order: a 1994 regulation ratified by the city people's congress (the city people's congress ruled that no firecrackers would be allowed); what appears to be a rather preachy authorial expression of opinion (However, the displays became not a pacifier but a fuse); other apparent authorial statements of fact that are actually drawn from a report in the Beijing Youth Daily, which is itself based upon a report from the Beijing Fireworks Ban Office; and statements by a generalized category of doctors, the authority for which also comes ultimately from the report in Beijing Youth News (see Table 7).

This depersonalization of negative evaluation is equally marked in China Daily text 42, where it is indirectly associated with two sources; the World Health Organization and Guangdong Province's poultry export agency, the Guangnanhang Company. The effect is to give an air of authority to the denials, to institutionalize them so that the reader feels he or she has little choice but to accept them. The effect of the personalization of negative evaluation as found in the Times, by contrast, is to heighten the sense of human drama, and simultaneously to decrease the authoritativeness of the denials (since a denial by a single, fallible human will always carry less weight than one by an institution the size of the World Health Organization).

Given the contrasting role played by the Times and China Daily in their respective societies, it would seem reasonable to predict that this pattern of personalization and depersonalization of negative evaluation would be repeated throughout the corpus of Times and China Daily texts. If so, we might expect to find that the proportion of negative evaluators that occur in the context of direct or reported speech is higher in the Times texts than in the China Daily.

The quantitative data summarized in Table 5 above revealed that, in general, comparator evaluation in the context of direct and reported speech was proportionally less likely to be found in the Times than in the China Daily. Specifically, just 49% of negative, future, and modal evaluation in the corpus of Times texts occurred in the context of direct or reported speech, compared to 60% per cent of that found in the China Daily texts. This seems counter to what might be expected on the basis of the discussion above. Break the data down, however, and the picture is different. In regard to negative evaluation, a significantly higher percentage occurs in the context of direct or reported speech in the Times (69% of all instances) than in the China Daily (43% of all instances) (Table 8). This would seem to support the conclusion that the pattern of personalization and depersonalization of negative evaluation noted above is repeated throughout the corpus of texts.

There is one other interesting aspect of the China Daily articles examined above. In each case, most instances of negative evaluation that occur appear to be used in the context of the narrative of the text itself; in other words, they appear to be in the journalist's own “voice.” Closer examination of the texts, however, reveals – as noted above – that while they may appear to be authorial statements of fact, in truth there is a specific authority given for these statements. China Daily article 42, about avian flu, is a good example. The litany of no cases appears to be simple statements of fact in the author's own voice, but in fact, it is clear that the information actually derives from the World Health Organization.

Future evaluators

As Table 3 shows, the Times texts analyzed use approximately twice as many negative and modal evaluators as the China Daily texts. Future evaluation, however, is different in that there is no significant difference in the frequency with which it is used in the two sets of texts (a future evaluator occurs on average once every 144 words in the Times, once every 155 words in China Daily). This is not, on the face of it, what might have been expected. So what is happening?

Future evaluators serve to place the narrative action – the actual sequence of events, taking place now or in the recent past, that are the subject of the narrative – in the context of events that have not yet happened but could happen in the future. They serve to contrast the now with the maybe, or the could be, or with what we want, hope, or expect might come to be.

In common with most other comparators, they heighten the interest of a narrative by placing it against a rich background of other events that have not actually happened, but could. It is not surprising, therefore, that they are one of the more commonly used forms of evaluation found in the Times texts. But why are they so widely used in the China Daily?

A closer examination of the China Daily texts that are richest in use of future evaluators gives some clues. These are China Daily article 2 with 11 instances, and China Daily articles 13 and 28 with nine instances each. China Daily article 2, headlined “One country, two systems,” is a clearly political piece stressing the importance of closer ties between the Chinese mainland and Taiwan. It is a comparatively long article, which might in part account for the number of future evaluators, but nevertheless, the way in which these evaluators are used is interesting. The instances of future evaluation found are set out in Table 9.

The future evaluators used here express plans, intentions, and hopes for the future. More generally, when taken together, they add up to a political vision of the future – a vision held by senior Chinese Party officials. If such and such happens, they say, then so and so will also happen, as a consequence of such and such or as part of a plan. The vision of the future held out here is implicitly contrasted with the present; implicit in the whole text is the sense that things are not, at present, the way the Party would like them to be. A better future will be the reward if certain things are done, runs the message.

China Daily article 13, headlined “New fares to encourage residents to take taxis,” is about a much less politically sensitive and important issue – promoting the use of taxis to take some of the strain off Shanghai's public transport system – but it still clearly encapsulates a vision of the future. Thus: Next month, taxis will charge ten yuan … . instead of 14.4 yuan; each kilometre after the first three … will be charged at two yuan; people who take a taxi seven kilometres or less will be paying slightly less; taxi drivers … say the change will hurt business; “The new pricing member will have a great impact on us,” said [taxi driver] Xiao Li; “I will earn just 30 yuan”; Bargaining will only result in unfair competition.

Here, a local government policy is being set out; we are being told simply what will come to pass. Interesting here is the fact that a voice of dissent is allowed to intrude: taxi drivers say the move will hurt their business, and that they will be worse off as a result of it. Such a voice of dissent is comparatively rare in the Chinese texts analyzed. It has perhaps been allowed here because the issue is not seen as a very important or sensitive one.

Finally, China Daily article 28, headlined “Chongqing plans ambitious move,” sets out another vision of the future – the city of Chongqing's plans to attempt to compete economically with Shanghai. Here we see future evaluators in clauses such as the following: Chongqing will provide more favourable conditions for foreign investors; “We will make Chongqing one of the best investment choices in China; more workers will be laid off from State-owned enterprises; the municipal government seems confident the problems will be solved. Again, intention and aspiration is the order of the day, with the repeated use of the future will implying a firmness and singleness of purpose that will brook no argument.

In the China Daily texts, then, future evaluators appear in the context of articles that set out official visions of the future. Doubts are occasionally allowed to creep in (taxi drivers fear they will be worse off; workers will be laid off), but the overall tone of the articles expresses firmness of purpose and planned progress, moving forward toward a planned better future.

A possible explanation for the comparatively frequent use of future evaluators in the China Daily texts, therefore, might be that the China Daily, in its role as mouthpiece of the Chinese Communist Party, is seeking in the articles in which such future evaluation is found to present the authorities' plans for a better future, and to portray the events of today as though they are taking place on the road toward a definite future goal.

This is reinforced by the fact that, unlike in the case of negative evaluation, the greater proportion of future evaluators in the China Daily texts occur in the context of either reported or direct speech; in other words, they are personalized. A total of 62% of instances of future evaluation found in the China Daily occur in the context of either reported or direct speech (see Table 10). What we are seeing, I would suggest, is senior party or other authority figures giving their views about the way forward.

China Daily article 2, discussed above, is a good example of this. Of the ten instances of future evaluation found, five occur in the mouths (in the context of either direct or indirect speech) of three named senior officials: Chinese foreign minister Qian Qichen; Chen Yunlun, director of China's Taiwan Affairs Office; or Tang Shubei, vice-chairman of the Association for Relations Across the Straits, Arats (see Table 9). Four more, though not all necessarily strictly in the form of reported or direct speech, are clearly associated with a shadowy authority referred to only as the mainland, which in the context of this article refers to the government of the People's Republic of China, and thus to the Communist Party. Only one is genuinely in the voice of the journalist: the brief statement It's still unclear when the semi-official talks will resume.

This personalization of future evaluation does not appear to occur to the same extent in the Times texts analyzed. While the frequency and spread of use of future evaluators in the Times appears, from the initial quantitative data, to be superficially similar to that in the China Daily, Table 10 reveals that, in contrast with the China Daily, only 15% of instances of future evaluation occur in the form of either direct or reported speech. The vast majority appear directly in the narrative.

The Times article displaying the greatest use of future evaluation is text 22, headlined “Middle class to foot bill for Budget reforms.” Here, future evaluators occur no fewer than 24 times. Of these, however, 20 are in the journalist's own voice. Just four occur in the form of readily attributable direct or reported speech (the one example marked * occurs in the narrative voice but can be reasonably attributed) (see Table 11).

Times text 22 is a preview of Chancellor Gordon Brown's first full budget, and the repeated use of futures is a result of predicting what the chancellor will do. It is certainly not a negative or hostile piece; rather, it is a speculative piece such as is not uncommon in the British press. Presumably the journalist has some grounds for writing what he or she has (possibly based on an advance copy or summary of the chancellor's speech, or on an off-the-record briefing by an aide, a supposition supported by two quotes directly from an unnamed source), but there is almost no attempt to attribute any of the information given. It is almost all delivered, quite flatly, in the reporter's own voice. The effect of this, I would suggest, is to distance the newspaper from Gordon Brown's plans, thus allowing it to stand back from them and assess or analyze them in what appears to be a more independent and objective manner.

The contrast with the way future evaluation is used in China Daily texts is clear. There, it occurs in the context of reported or direct speech attributable mainly to named individuals or organizations. The effect is for those organizations or individuals to be able to claim ownership (and hence credit) for the plans or intentions being reported, and at the same time to make the newspaper seem supportive of them. The Times, by reporting such plans in the form of simple statements of fact in the author's own voice, actually manages to sound more distant and skeptical about them, and is able to place them in the context of an article which appears to be weighing the merits of the plans in the balance.

The same is true of other Times texts. The one showing the second greatest use of future evaluation is text 50, headlined “Tax credit for poorer families will go to women,” which contains 11 instances. They include: A new tax credit to top up the income of lower paid families will now go automatically to women; couples will fill out a joint form, which will include a box which women can tick; the new credit will also include a special payment for childcare; It will be more generous than the family credit. Again, we see little or no attempt to place the facts stated in the voice, direct or indirect, of any government official or spokesman, allowing for a more distant, skeptical, and analytical article.

Modal evaluators

The third type of comparator being looked at here is the modal evaluator. In traditional grammar, modals in English mainly indicate the attitudes, abilities, or opinions of the subjects of a particular clause toward the actions or states described in that clause. These attitudes or abilities may include willingness (I would love to go if the weather's OK); ability (he can read: I can't); obligation (you must go); permissibility (you may go); possibility (you might be able to get there that way); and volition (I will go!). Here, though, we are examining the role of modals as comparators: deliberate linguistic choices made by the narrator to enrich the actual sequence of events reported on by setting them against a wider background of things which may, or could, or will, or must, or can't happen.

A look at the quantitative data reveals that the Times articles analyzed use over twice as many modal evaluators (188) as the China Daily articles analyzed (78) (see Table 1). Again, this is not principally a function of the longer length of the Times articles. Even when the slightly shorter average length of the China Daily texts is taken into account, modal evaluators appear twice as frequently in the Times texts analyzed (on average once every 106 words) as in the China Daily texts (once every 213 words) (see Table 3). As we saw, these differences are highly significant. Use of modal evaluation is also more evenly spread through the corpus of Times texts than is the case with the China Daily texts. 88% of all Times texts analyzed contain a modal evaluator, compared with just 42% of China Daily texts; the top five Times texts contain 32% of all modal evaluators found in the corpus, compared to 46% in the top five China Daily texts (see Table 4).

Clearly, the pattern and frequency of use of modal evaluation is quite different in the Times texts and China Daily texts analyzed. But why?

The Times article showing the greatest use of modal evaluation is Times article 31, which contains 15 instances. Headlined “Outside agencies may investigate police complaints,” the article is an account of proposals to take away responsibility for investigating complaints against the police from the police themselves and instead to appoint independent investigators. Modal evaluation found is set out in Table 12.

Here, as was the case with future evaluation, we are looking at a text that is talking about a vision of a possible future. Implicit in the story is a contrast between the situation now – in which the police themselves handle complaints made against them – and what the situation could be in the future, with those complaints handled by external investigators. The difference between the vision of the future expressed here and the visions of the future expressed using future evaluation in the texts analyzed earlier is that here we are talking not about definite plans for the future, but about possibilities – something that participants would like to bring about, if it were possible. There is a lesser degree of certainty implied in this possible vision of the future than if future evaluators had been used. If, for example, Jack Straw had said Inquiries will be taken away from police forces, that would have been a definite intention, a planned future move. By saying they could be taken away, he is talking about something that may happen, but also may not.

Times article 46, headlined “Irvine calls for curbs that would suppress Cook story,” shows the second highest usage of modal evaluators (13 instances). This is an article looking at the impact of new curbs on press freedom being called for by the Lord Chancellor, Lord Irving. Modal evaluators include the following: Robin Cook's affair with his mistress … would not have been disclosed under new curbs; people could go to the commission and ask it to stop stories being published; In an interview … Lord Irvine is asked if he would have expected the commission to order the News of the World not to print the story; “I would hope that that would be the view that the PCC would form in a case like that,” he said.

Here, again, we have a vision of a possible future that is contrasted sharply with the reality of the today of the article. Robin Cook's affair was in fact reported, but it would not have been if the suggested new curbs on press freedom had been in place. Again, though, the use of modal evaluators lessens the sense of certainty or inevitability that this vision of the future will come about. People will be able to go to the commission would have implied this will be the case; people could go to the commission implies it may be the case if something else also comes to be – here, clearly, if the Press Complaints Commission issues tough new privacy guidelines. In this article, the lack of certainty appears to derive from the fact that it is not Lord Irvine himself who has the power to issue such guidelines. He can only call upon the Press Complaints Commission to do so.

Times texts 19 and 30, both comparatively rich in use of modal evaluators, show a similar pattern. Text 19, headlined “Supermarket and hospital car parking may be taxed,” is an account of government proposals to levy a tax on parking spaces, but these are only measures being considered. Thus, we read The money would probably be levied voluntarily and the proposed charge could raise £650 million. Times article 30, headlined “Police chiefs get power to sack corrupt officers,” is a story about definite powers to be granted by Home Secretary Jack Straw. It does therefore include future evaluators (The Home Secretary will announce the biggest shake-up of the handling of police complaints). Where the modals begin to appear is when the journalist begins to speculate about precisely what the shake-up will mean: If officers refuse to answer questions this could be used against them; Officers who are lazy or incompetent could also find themselves facing the sack. In the Times articles, then, modal evaluation clearly is being used primarily to describe a vision of a future, but a possible, or wished-for, or speculative vision of the future rather than a planned, intended one as is the case when future evaluation is used.

Modal evaluation occurs considerably less frequently in the China Daily texts than in the Times texts, but it does occur. A closer look at the China Daily text displaying the greatest use of modal evaluation – article 2, headlined “One country, two systems,” about the importance of improved business and economic relations with Taiwan, which contains 10 instances – reveals, however, that this device is not being used in the same way as in the Times. China Daily article 2 – also the Chinese text that displays the greatest use of future evaluation – clearly sets out a vision of a better future. But the modal evaluators found tend to be those of obligation, rather than of possibility as in the Times texts (see Table 13).

All but one of the modal evaluators found in China Daily article 2, in fact, is a modal of obligation such as must, can, or should; and, as Table 13 shows, most of them are directly attributable to a senior government source, China's then vice-premier and foreign minister, Qian Qichen.

The contrast with the type of modal evaluation found in Times article 31 is interesting. As with the China Daily text, most of the modal evaluators found in Times article 31 are attributable to a senior government source, in this case Jack Straw, then home secretary. But the types of modals used by Straw are almost exclusively modals of possibility or speculation. Straw uses only a single modal of obligation (bad officers must be dealt with) and the Commons Select Committee on Home Affairs another (the Home Office should look at) (see Table 12).

This difference in types of modal evaluators possibly says as much about the relationship between the government and the governed in the two societies as it does about the roles of the media. The pattern of China Daily texts using modals of obligation and Times texts using modals of speculation is, however, repeated throughout the corpus of texts. Of the 74 modal comparators found in the China Daily texts, 52 (70%) are modals of obligation such as must, can, or should. In the Times texts, just 30 out of 186 (16%) of modal comparators are modals of obligation. The vast majority of modal comparators in the Times texts (145/186, or 78%) are those of possibility or speculation (would, could, may, might). This compares with just 22/74 (30%) of the modal comparators in the China Daily texts.

Given the probing, speculative nature of much Western reporting, it might be expected that one reason for the large proportion of modals of speculation found in the Times corpus is that they occur in the context of the narrative and are the result of the journalist's speculating on, for example, what certain government proposals might mean. There are indeed a couple of examples of this kind of use in Times article 31 (new legislation … would take some time to arrange). By and large, however, as Table 14 reveals, it is not the case that significantly more modals of evaluation occur in the context of narrative as opposed to direct or reported speech in the Times than in the China Daily. In the Times, 38% of modal evaluators occur in the context of narrative; in the China Daily, this is only slightly lower, at 33%.