|C R I T I C I S M . C O M :: Home Media Criticism newslang||Search | Site Map | About Criticism.Com | Home|
On the Wire
Journalism In the News
No current headlines.
Sociolinguistics of the News Media: Motivations for News Language Style -- Audience Perception or Cultural Orientation?
By Steve Hoenisch
Last updated on March 1, 2006
Copyright 1996-2006 www.Criticism.Com
Table of Contents
2 A Basis for Comparison
3 "Radio's" Explicit and Implicit Claims
3.2 Negative Contraction and Institutionalized Style
3.3 Style Shift
3.4 "Radio's" Conclusion and Its Conflict with that of "Media"
4 The Contentions of "Media"
4.1 Problems with the Rule
4.2 Where Have all the Determiners Gone?
4.3 The Lost Comma
4.4 The Social Force of Determiner Deletion in U.S. Print Media
4.5 The Heart of "Media's" Findings
4.6 Historical Development
This essay examines the methodology and findings of two of Allan Bell's related sociolinguistic studies on the language of the news media: "Radio: The Style of News Language,"1 henceforth "Radio," and "The British Case and the American Connection in New Zealand Media English,"2 henceforth "Media."
In particular, the essay will argue that, given the similarity of the linguistic variables Bell is analyzing in the two studies, his hypothesis of audience influence upon language in "Radio" at least partially conflicts with his hypothesis of cultural orientation in "Media" -- that is, both hypotheses cannot be entirely correct unless one is clearly subordinated to the other, a move that Bell does not make in the later article.
Along these lines, I will argue that the explanation in "Radio" seems better supported by empirical evidence presented not only in the article but also in Chapter 6, titled "Stylin' the News: Audience Design, in Bell's The Language of the New Media, while the explanation in "Media" is ad hoc and lacks convincing empirical support.
Bell would have done better to apply "Radio's" hypothesis of audience perceptions to the data in "Media" instead of maintaining a hypothesis of cultural orientation. The first course is stronger because, to the extent that the cultural orientation hypothesis is appropriate, it is parsimoniously included in that of the audience perception hypothesis. Yet without the explicit subordination of one hypothesis to the other, the two hypotheses stand in opposition, both vying for the dominant level of explanation.
The essay will also look at Bell's methodology and the veracity of the conclusions that can be drawn from it. Furthermore, the essay will also take a general look, in passing, at the substance and sociolinguistic relevance of Bell's studies, noting on a broad level how the studies are indicative of wider problems within sociolinguistics and noting on a micro level several possible shortcomings in Bell's methodology.
First, I will look broadly at the content of "Radio" and summarize the claims of "Radio" as well as examine its methodology. I will next do the same for "Media." In the course of examining the two studies, I will point out why the factors that are said to influence language usage conflict.
Before proceeding to look at "Radio," I would like to show why both studies can be reasonably, though perhaps not precisely, compared on the same footing: Some of the linguistic phenomena in the two studies are adequately similar to allow such a comparison. In "Radio," the three linguistic variables that Bell investigates are negative contraction, consonant cluster reduction, and intervocalic /t/ voicing in radio news speech. In "Media," Bell analyzes rates of determiner deletion in the language of newspapers and radio news. In radio news language, all four variables come into play and can be easily analyzed as similar phenomena. In the print language of newspapers, on the other hand, neither intervocalic /t/ voicing nor consonant cluster reduction come into play, but negative contraction and determiner deletion are both present and easily grouped as complementary variables -- both indicate the use of an informal style.
A further indication that the variables are similar enough to allow comparison is that Bell goes on to use the determiner deletion data from "Media" to support an audience perception hypothesis in Chapter 6 of The Language of News Media.
"Radio's" descriptive content and its findings are characteristic of much of the work that underlies sociolinguistic research: Painstaking confirmation of the obvious. A lay observer of language knows intuitively, for instance, that -- and this is characteristic of the nature of Bell's findings -- a high rate of negative contraction typifies the language used on rock radio stations, even during their news segments, as these stations are obviously inclined to be more informal than, say, stations broadcasting classical music. A high rate of negative contraction is simply a characteristic of informal speech and would thus quite obviously be present in broadcasting formats aimed at less formal audiences, such as teenagers and young adults.
More interestingly, "Radio" also examines the factors that influence the language style of news. The study lays down a strong claim: A radio station's perception of its audience's social status influences the style of language in which the station presents the news. But there is also an implicit claim in Bell's study: That newscasters' adaption of language style rests on their perception of audience provides indirect evidence for the hypothesis that speakers in face-to-face communication alter their language depending on whom they are talking to. In his book The Language of the News Media, Bell makes the claim explicit: "The processes which mould language style in mass communication are similar in kind -- but often greater in degree -- to those which operate in face-to-face interaction,"3 a claim I believe to be fundamentally flawed.
My first objection to the claim that newscasters adopt their language to whom they perceive to be their audience is that there are other factors that influence the kind of language used in radio news: the subject and content of the news itself (that is, whether it is serious or grim, familiar or light, local or distant, etc.); the type of music that the station plays; and the station's nonmusic programming. Although some of these variables are acknowledged by Bell and held constant in the study, others, such as music programming, do not seem to be given adequate weight. More important, however, is the influence on media language style of institutionalized rules that do not affect the style of face-to-face communication. People simply do not implement an explicit style for dealing with all their possible audiences.
In fact, overlooking the influence of institutionalized style rules on media language points to a crucial error by Bell, one that will come to the fore in this essay: He confuses the motivations for the style used in personal, face-to-face communication with the motivations for the style used by the media. In The Language of the News Media, Bell writes: "I believe the essence of style is that speakers are responding to their audience. It is typically manifested in a speaker shifting her style to be more like that of the person she is talking to,"4 entailing that each listener may be spoken to differently. Whereas individual speakers may respond to the person to whom they are speaking, the media can respond only to the aggregation of whom they believe to be their audience. Style guidelines are established to ensure that writers address their audience in the same way in every story over time. One fundamental reason this crucial difference exists, besides that a medium often does not know exactly who its audience is, is that the makeup of a medium's audience may be quite diverse and vary greatly over time.
At any rate, it is clear that the language of news radio is likely to be influenced by an interrelated web of factors, some of which may be independent of audience, such as institutionalized style in the way of rigidly followed stylebook guidelines. To hold that one and only one factor -- perception of audience -- directly influences the language style of newscasts seems dubious.
The same objection holds for face-to-face communication: People vary their speech according to a variety of interacting factors, some of them independent of audience. These factors include, among others, the feelings and mind set of the speaker as well as the setting, time, and situation during which the communication is taking place.
Moreover, some cultural factors that affect face-to-face speech, especially in diverse immigrant communities like New York, may lie beneath the observable surface, masked by acculturation or superficial attempts at assimilation.
Regardless of such extra-audience factors as the influence of culture in interpersonal communication, however, there are other reasons the methodological approach that uses media language to explicate difficult-to-study speech patterns is misguided. The most important is that the language of the news media -- which in the case of radio news is usually based upon written text -- may not be a reliable indicator of how different sets of people speak. In other words, formulating conclusions about how a radio station's audience speaks based on the broadcasting speech used by radio station announcers is methodologically dangerous, and likely to result in, at best, imprecise findings. Written language, even when modified for oral presentation on the radio, is often significantly different from spoken language. The researcher better serves his interests and those of sociolinguistics if he or she attempts to find a direct approach to studying the phenomena in question.
A bigger methodological problem emerges when Bell goes on to analyze the three linguistic variables of negative contraction, consonant cluster reduction, and intervocalic /t/ voicing. The problem, I would like to suggest, is that identifying rates of change among stations with respect to these three linguistic variable does nothing to sufficiently pinpoint the cause of changes. That is, changes in these three linguistic variables could be attributed either to an attempt by a station's news broadcasters to mold their style to their perceived audience or to the station's attempt to attract a certain audience to its established programming. Indeed, the latter possibility, rather than the former, could be borne out by Bell's finding that the news language style of different newscasters at the same station did not vary, revealing a strong possibility that the station's newscasting style is institutionalized. At any rate, it remains unclear whether the style is institutionalized to serve what it perceives to be its audience or to attract and retain a particular audience.
Another possible problem with attributing the variation of radio speech to announcers' perceptions of the social status of their audiences is the lack of extensive empirical data and confirming research on the social stratification of New Zealand. Bell acknowledges that the educational and occupational scales in his study are based on a rather slim collection of literature, but fails to examine cases in which the dearth of data could lead to misinterpretations of an audience's status. In "Radio," Bell draws largely on a single source, an unpublished Ph.D. manuscript5, for his social stratification data, leaving much of the data in his study without independent corroboration. Thus, because there is so little data on social divisions in New Zealand, conclusions based on this data could turn out to be inaccurate, or worse, wrong. Furthermore, because six years had passed between the presentation of the data in 1968 and Bell's 1974 study, Bell's findings tied to social status may have already been slipping out of date.
Another possible explanation that Bell overlooks for some of the linguistic variations studied in "Radio" are those that may be shaped by a radio station's stylebook or overt policies on broadcasting speech. For instance, Bell says a striking feature of the BBC newscasters' is that they never contract negatives. Such a pattern, however, is probably better attributed to the influence of a set broadcasting policy than to the broadcasters' perception of their audience. And if it is a policy that the BBC does not contract negatives in news broadcasts, than such a pattern is less striking than otherwise. The New York Times, for instance, has an institutionalized policy on negative contractions. Its stylebook says: "In ordinary news copy, spell out such expressions as is not, has not, have not, do not, are not, will not, etc. But contractions are acceptable in quoted matter, in feature contexts and in headlines."6 The guidelines published for its editors by the Associated Press, which prepares news copy for distribution to radio stations as well as newspapers, takes a somewhat softer stance but nonetheless advises its editors to "avoid the excessive use of contractions."7 If the policy on contractions of the BBC and the other stations in Bell's study resemble an institutionalized policy like that of The New York Times or the Associated Press, negative contraction should be thrown out as an uninformative linguistic variable. In fact, a station's overt policy on contractions would reduce them from the status of a sociolinguistic variable to a highly predictable component of their speech. After all, a sociolinguistic variable, a methodological construct developed by William Labov, "is a set of alternative ways of saying the same thing, although the alternatives will have social significance."8
An institutional policy on the use of negative contractions would remove the possibility of an alternative way of saying the same thing. Bell, before using negative contractions as a sociolinguistic variable in his study, should have determined the extent to which, if at all, each radio station had a overt policy governing their use.
Consonant cluster reduction and intervocalic /t/ voicing, however, are better indicators of speech patterns, for they are far less likely than negative contraction to be regulated by a style policy.
In "Radio," Bell writes: "Single newscasters heard on different stations show a remarkable and consistent ability to shift their style considerably to suit the audience," and his data, even after subtracting negative contraction, bears this out. The same four newscasters, who alternated between two government-operated stations, varied their intervocalic /t/ voicing typically by 15 percent, depending upon the station over which they were broadcasting. These findings, coupled with data showing that there was little to no variation among individual newscasters at the same station, Bell's says, strongly support the hypothesis that newscasters vary their language style depending upon who they think is listening.9
In "Radio," Bell concludes that "newscasting seems to be, then, a case where the speech patterns of individuals are subordinated to corporate style, which correlates only with audience characteristics."10 Besides the indeterminability of cause pointed out above, the problem with this finding is that it clashes with Bell's findings about determiner deletion in "Media." Corporate styles of determiner deletion, as presented in "Media," correlate not with audience characteristics, as Bell's conclusion of "Radio's" should have suggested, especially given the similarity of the linguistic variables between the two studies pointed out above, but are found to be determined by cultural orientation. Both hypotheses can hold only if cultural orientation is explicitly subsumed as a characteristic of the audience, but Bell, unfortunately, does not attempt to make this connection in "Media." Bell, in "Media," which was written after "Radio," should have directly pursued the possibility of this causal connection rather than merely stopping with the vague and suspect cultural orientation hypothesis.
In "Radio," moreover, Bell says that "in mass communication, the pressure to seek approval is at its maximum, since media live by the size and composition of their audiences." This assertion, if correct, and Bell believes that his findings in "Radio" bear it out, leaves unexplained the shifts within the New Zealand news media toward American-style levels of determiner deletion -- unless those shifts are somehow tied to an attempt by the media to seek audience approval and retention. Yet the latter possibility can only be sustained if audiences are seen as moving toward American culture. In other words, if Bell's conclusions are correct in "Radio," higher rates of determiner deletion reflect changing perceptions by the media of their audience's cultural orientation. The rates of determiner deletion do not, however, necessarily reflect an orientation of the media toward American styles independent of perceptions of audience characteristics.
As I've already mentioned, Bell puts forth a different hypothesis in "Media" than in "Radio" to account for the behavior of a similar linguistic variable. He writes: "How frequently New Zealand mass media apply the rule [of determiner deletion] proves to be diagnostic of their orientation towards the culture and language of one or other" of the United States or Britain.11 "Use of the rule of determiner deletion," Bell concludes, "therefore exemplifies a wider orientation by New Zealand media to the international models of English and the cultures from which they come."12
Unfortunately, a lack of examples with their complete context makes it more difficult than it ought to be to analyze Bell's syntactic rule of determiner deletion and its scope in "Media." "The rule," Bell says, "deletes the determiner in appositional naming expressions of the form (the) race relations conciliator Hiwi Tauroa; (a) local resident Beth Anderson; (his) `Chips' series co-star Erik Estrada."
There are problems with the way the rule is presented. First, such examples as the second one above -- (a) local resident Beth Anderson -- would never be printed in American media without the name being set off by commas or without the determiner deleted. That is, in American print media, it would appear either as
(i) "Local resident Beth Anderson" was on the board of education"
(ii) "A local resident, Beth Anderson, was on the board of education"
but never, even in The New York Times, which Bell says has one of the lowest rates of determiner deletion among daily newspapers in the United States, as
(iii) *"A local resident Beth Anderson was on the board of education."
Bell is directly concerned with appositional determiner deletion, but because of the above usages, a revealing study of determiner deletion in American newspapers would look more closely at post-appositional deletion, where
(A) "Beth Anderson, a local resident, was on the board of education"
is the common form used by broadsheet dailies like The New York Times, while
(B) "Beth Anderson, local resident, was on the board of education"
typifies the post-appositional style used by tabloid dailies like the Boston Herald or the New York Daily News and by broadsheets like The Salem Evening News in Massachusetts that perceive their audience as being predominantly working class. In the United States, most daily tabloid newspapers, with the exception of New York Newsday and, perhaps, The Rocky Mountain News, are oriented toward working-class readers while their broadsheet counterparts appeal to a higher class.
A sociolinguistic study of American determiner deletion such as Bell's that looks at appositional deletion patterns alone while discarding post-appositional patterns lacks the scope to ensure that its findings indicate actual patterns of linguistic variation in American news language, not merely a confirmation of a rather obvious grammatical rule of American English, as exemplified by (i), (ii), and (iii) above. In British English, "A local resident Beth Anderson" is a permitted form, used by the linguistically conservative broadsheets, while in American English, without the name set being set off by commas, it is not. The deletion of indefinite determiners in such forms as (i) is likely to be near 100 percent in American newspapers, with perhaps a percentage point or two of insignificant variation attributable to editing and production errors.
Considering that (iii) is an illegal form for American newspapers, both tabloid and broadsheet, and a permitted form for British newspapers, used by U.K. broadsheets but not by tabloids, a study that uses determiner deletion rates in appositional phrases with an indefinite article to find that New Zealand are oriented toward American print media will misdiagnose the orientation of the New Zealand media. The findings would not imply an orientation toward American media, but only an orientation toward the language style of British tabloid media, which, as we will see, may or may not have been influenced by U.S. print media to begin dropping determiners. All tokens of the form (i), with the indefinite article, must be discarded from the American sample. And this Bell does not do.
There are similar, though less severe, problems with the definite article in appositional position. The stylebook of American Banker, a prestigious daily financial newspaper that operates in the linguistically conservative tradition of American broadsheets, gives the following rule: "Descriptions other than genuine titles should not be affixed to the name (Merrill Lynch chairman James Jones) but either placed after the name (James Jones, chairman of Merrill Lynch) or offset with a comma before the name (the chairman of Merrill Lynch, James Jones, ... )."13 Similarly, the stylebook of The New York Times advises its writers and editors that "long titles should follow names." It goes on to say: "Only genuine titles -- not mere descriptions, whether lowercased or capitalized -- should be affixed to names. Do not, for example, write harpsichordist Joan Manley or Political Scientist John P. Manley. But a phrase in apposition, preceded by the, is acceptable: the sociologist Margaret Manley."14 While acceptable, however, the latter construction tends to appear only rarely in American print media. Preferred constructions are, as the New York Times Stylebook points out, those wherein the name following the appositional phrase is set off by a comma: "the majority leader, John P. Manley."15 As mentioned above, an appositional phrase beginning with an indefinite article -- "a sociologist Jane Manley" -- would never appear without being considered an error.
Thus, with respect to an example that Bell uses, only (D) would arise at American broadsheets without breaking the style rule:
(C) *fugitive financier Mr. Robert Vesco (D) the fugitive financier Mr. Robert Vesco
Yet (C) and (D) nevertheless differ in an important way. In the language of many American print media, especially tabloids, (C) is a possible, popular, and, unlike at American Banker and The New York Times, permitted form while (D) is an unpopular though acceptable variation. Thus, in general, the legal possibilities would be
(E) Mr. Robert Vesco, fugitive financier
(F) Mr. Robert Vesco, the fugitive financier
(G) the fugitive financier, Mr. Robert Vesco
(H) the fugitive financier Mr. Robert Vesco
(I) fugitive financier Mr. Robert Vesco
The construction shown in (D) and (H), however, would not appear without being considered at least unusual, as is the case at The New York Times, or as an error, as would be the case at nearly all other newspapers, including not only tabloids like the New York Daily News and the Boston Herald but also such broadsheets as The Hartford Courant and American Banker.
On the other hand, (I) would not appear without being considered an error at such broadsheet dailies as The New York Times, The Washington Post, and American Banker.
Yet, in a large sample from The New York Times, Bell reports a determiner deletion rate of 74 percent, significantly and surprisingly high for a linguistically conservative U.S. broadsheet given the nature of the rules outlined above. In the context of the above rules, The New York Times should be expected to have a very low rate of determiner deletion. Where are all the determiners going?
In a similar vein, the rate of determiner deletion was even higher for The Washington Post; it was 91 percent. How can The New York Times's and The Washington Post's high rates of determiner deletion be explained given the infrequent use of (C)? Again: Where have all the determiners gone? Why do the two American broadsheet dailies that Bell cites -- The Times and The Post -- have such high rates of determiner deletion? After all, the high rate does not seem to stem from the appositional deletions of definite articles.
Bell's determiner deletion data also include indefinite articles and possessive pronouns. Assuming, then, that the high rates among U.S. broadsheet newspapers do not stem from the deletion of definite articles, the answer must lie in one of the other two variables that Bell uses: indefinite articles and possessive pronouns. The likely explanation, then, as touched on above, is that Bell is counting instances of (i) as determiner deletions when they should in fact not be counted as such, since their pre-deletion state, (iii), is illegal. In such cases, there is never a determiner to be deleted. (In this regard, Bell is not distinguishing between phrases set off by commas and those not set off by commas, a crucial distinction that must be made to properly analyze the use of determiner deletion given such rules as those outlined above.)
Another possibility, though, is that deletion of possessive pronouns in the appositional determiner position is uncharacteristically high, significantly bringing up the rate of determiner deletion. This, however, seems unlikely, but given the available data, such a possibility can be neither affirmed nor denied.
But if we assume that the appositional deletion of possessive pronouns is neutral -- that is, neither outrageously high nor significantly low -- then Bell's data inflate the rate of determiner deletion among U.S. broadsheet newspapers. Further study would be required to tease out the possible influence of the deletion of possessive pronouns. Bell should have distinguished, as he did with the samples from British media, between the deletion of all determiners and the deletion of articles only in his data on the U.S. media. Such a distinction would have allowed a better interpretation of the U.S. data. Likewise, he should have also distinguished between indefinite and definite article deletion.
Another, related problem emerges in the categorizing of radio media language, for a distinction between (D), the highly unusual and infrequent form, and (G), the popular form, can not easily be made by sampling radio speech, and indeed Bell opts not to make it. But, I believe, this further skews the data, resulting in an inaccurate characterization of determiner deletion.
From a passage in "Media," it seems that Bell may indeed be dropping the all-important placement of the comma in his samples, but relevance of the passage to the treatment of appositional samples is unclear because it pertains to post-appositional forms: "When the name NP precedes, this particular rule is inapplicable, although the determiner may still be deleted from the following descriptive NP: as in Mr. Robert Vesco (the) fugitive financier."16 In American print media language, however, such a form, with or without the determiner, would never intentionally appear without a comma after the last name.
Bell's rule of determiner deletion makes no mention of comma placement or use, though he does go on to say that "the written form invites a comma between the two NP's of the apposition, marking the parenthetical nature of the second NP."17
For Bell's study to have descriptive adequacy, especially with respect to the print media samples, the use of commas must in some way be taken into account, though this obviously presents problems for collecting samples of radio language.
In commenting upon the U.S. data, Bell holds that the low spread of differences in the rates of determiner deletion -- that is, semicategorical deletion -- signifies that the rule has little social force in the United States. However, my intuition, coupled with my extensive knowledge of U.S. print media language acquired from working for 7 years as an editor at daily newspapers, both broadsheet and tabloid, both prestige and popular, is that there is a social difference that can be detected among American print media if the syntactical differences in the determiner deletion rule for American media and British media is adequately captured and if post-appositional determiner deletion is taken into account. Much of this difference has already been exposed above, in evaluating constructions that appear predominantly in broadsheets, those the appear mostly in tabloids, and those that readily appear in both. Specifically, the difference is this: In the United States, the more prestigious broadsheet newspapers -- like The Boston Globe, The Washington Post, The Hartford Courant, and The New York Times, which appeal to the higher social classes -- adhere to a style rule similar to that of American Banker's, excerpted above. That is, they conventionally use a form of nondeletion in appositional position but with the name set off in commas:
(X) The chairman of the board, Michael Milken, ...
and, less frequently, the form
(X') Chairman of the board Michael Milken ...
On the other hand, the tabloid newspapers, which generally appeal to lower social classes than the broadsheets, use the form (X') as their standard and the form (X) much less frequently. (The form (D), as I've mentioned, is unusual though permitted by The New York Times but illegal at such other prestige newspapers as American Banker.)
In post-appositional position, the social force of determiner deletion is even more strongly marked. The broadsheets would conventionally use the form
(Y) Michael Milken, the chairman of the board, ...
and less commonly the form
(Y') Michael Milken, chairman of the board, ...
while tabloids would conventionally use (Y') and rarely (Y). Indeed, it is in post-appositional position that the social force of determiner deletion makes itself felt in U.S. print media.
It should be noted, however, that the broadsheets are probably increasingly moving toward the determiner deletion standards of the tabloids, with the changes apparently stemming from the writer-reporter side of the news production process, with editors attempting to hold the fort against the onslaught, winning battles but ultimately losing the war. Perhaps it would prove worthwhile to carry out a study that attempted to pinpoint who in the production process of the news is having the most significant impact on determiner deletion: the reporter, the copy editor, the main editor, or, in radio, the announcer.
The heart of Bell's sociolinguistic findings on the application of the determiner deletion rule in "Media" falls into three categories: geographical differentiation, social variation, and historical development.
Geographically, the findings indicate, Bell says, that "there is a clear polarization between media in Britain and in the United States."18 Yet the findings regarding the U.S. print media are inadequate in showing, as Bell contends, that determiner deletion has little social force. First of all, and importantly, there are no tabloids in the U.S. samples. Second, the comparatively low level of determiner deletion in The New York Times -- 74 percent compared with The Washington Post's 91 percent and the two TV stations sampled, ABC at 92 percent and CBS at 90 percent, signifies the possibility that the rule has social force. It also signals the need to further study determiner deletion in U.S. print media before concluding that the rule has no social force.
With respect to social variation, Bell finds that the extent of determiner deletion in both British and New Zealand media corresponds to the social status of their readerships. Specifically, Bell finds that "the ranking of [British] newspapers ... for their degree of determiner deletion corresponds almost exactly to the social status of their readerships,"19 with high rates of determiner deletion corresponding to lower social status and low rates corresponding to higher social status. My only methodological concern regarding these findings is that Bell makes no distinction between determiner deletion rates for indefinite and definite articles, a distinction that might be relevant, for reasons stated above, in comparing British media to U.S. media.
Bell finds the same correspondence between rates of determiner deletion and audience social status for New Zealand radio stations as he did for British newspapers. Commenting on this, he goes on to say, "The polarization of radio stations reflects very accurately their orientation towards British or American cultural and linguistic norms."20 Further: "The rule of determiner deletion turns out to be diagnostic of New Zealand media orientations, just as it was for British media."21
Regarding historical development, Bell finds that between 1974, the time of his first sample, and 1984, the time of his second sample, New Zealand radio media showed a large shift toward the higher levels of deletion characteristic of U.S. media. Two of the four stations studied, in fact, underwent "a rapid and massive leap in the normally slow timetable of linguistic change,"22 Bell writes. The sample from one of these stations, however, was small compared with the others: only 35 compared with a range of 123 to 222 for the other three stations. Bell also looks at changes in the rate of determiner deletion for a British tabloid, The Daily Mirror, between 1920 and 1980, finding that it had a determiner deletion rate for articles of 0 percent in 1920 but a rate of 94 percent in 1980. Bell attributes historical changes in determiner deletion among the British press to American influence. "The mass press" in Britain, Bell writes, "has gone over to what it sees as the less formal, more popular American style using determiner deletion."23
Although it may be justified and correct in the end, there are several questions left unanswered in making the inference that American media have influenced the British tabloids to drop determiners.
First, if as Bell claims there is no social force behind the U.S. media's use of determiner deletion, why would the British tabloids but not the broadsheets adopt the American style? Bell's answer is that the alleged American style of high deletion is seen as "less formal, more popular" than the British style of less deletion. Within the data of Bell's study, however, the problem with this association is that it finds the British tabloids following the usage of the American broadsheets just as much as that of the American tabloids, and there is no indication that The New York Times and The Washington Post can be seen as using a "less formal, more popular style," at least in the context of American English. In fact, in the context of American journalism, they are seen as using a formal, conservative, and prestigious style.
Another concern bears on the nature of using cultural orientation hypothesis, showing how convoluted its application can become. Can the British tabloids not be seen as reorienting themselves toward American culture? Bell does not explore this possibility. At any rate, the possibility seems dubious, given that the British tabloids have long influenced American tabloids, even to such an extent that, traditionally, mass-circulation East Coast tabloids like the Boston Herald and the New York Post have imported British editors to fill high-ranking positions.
A second, more significant problem is that, for all we know, determiner deletion could be the same figure for America broadsheets in 1945 as it is for British tabloids during the same year. On such a possibility, attributing adaption of determiner deletion to the influence of American newspapers is unsupported. Indeed, this possibility becomes all the stronger when the incremental changes in British tabloid use of deletion are examined over time. The largest change -- 32 percentage points for articles -- takes place between 1940 and 1950. The percentage change for determiner deletion decreases every decade thereafter. Between 1950 and 1960, it is already only 17 percentage points, and between 1970 and 1980, it is a similar 14 percentage points, showing that if the British tabloids have in fact adopted an American style, they did so decades ago, mostly in the years between 1940 to 1950. Yet now we need to know what the rates of determiner deletion were in the 1940s in the U.S. media. Was the rate higher or lower than that of British tabloids by 1950 -- 47 percent? In light of Bell's own numbers, then, the history of American determiner deletion must first be examined before it can be said with any certainty that the British tabloids have adopted an American norm.
Likewise -- and with this remark I will begin to conclude my essay's argument -- it now becomes possible that the New Zealand media are adopting not an American style but the style of the British tabloid press.
And there is yet another possible explanation: Tabloid newspapers, because of their tighter news holes and tendency toward presenting shorter, pithier news stories, also have a natural, independent motivation to adopt a style of determiner deletion, which condenses the news language. Thus it is possible that the New Zealand newspapers have merely moved quickly to adopt the style of tabloid newspapers in general in order to save space, not to adjust to what they perceive as changes in their audience's cultural orientation.
In the end, then, Bell's concluding explanation in "Media" that the "use of the rule of determiner deletion therefore exemplifies a wider orientation by New Zealand media to the international models of English and the cultures from which they come"24 is neither sufficiently supported to exclude alternative explanations nor wide enough to accommodate internal discrepancies in the application of Bell's rule of determiner deletion.
1. Bell, Allan. 1982. "Radio: The Style of News Language." Journal of Communication 32: 150-64.
2. Bell, Allan. 1988. "The British Base and the American Connection in New Zealand Media English." American Speech 63.4: 326-44.
4. Ibid. p. 105.
5. Vellekoop, C. "Social Stratification in New Zealand." Unpublished Ph.D. dissertation, University of Canterbury, New Zealand, 1968.
6. Jordan, Lewis ed. The New York Times Manual of Style and Usage: A Desk Book of Guidelines for Writers and Editors (New York: Times Books, 1976), p. 49.
7. French, Christopher W. ed. The Associated Press Stylebook and Libel Manual (New York: The Associated Press, 1987), p. 52.
8. Fasold, Ralph. The Sociolinguistics of Language (Oxford, U.K.: Blackwell, 1990), p. 223-224.
9. Bell, Allan. "Radio," p. 163.
10. Ibid. p. 163.
11. Bell, Allan. "Media," p. 326.
12. Ibid. p. 327.
13. American Banker online stylebook. Unpublished. 1994.
14. Jordan, Lewis ed. The New York Times Manual of Style and Usage: A Desk Book of Guidelines for Writers and Editors (New York: Times Books, 1976), p. 208.
15. Ibid. p. 122.
16. Bell, Allan. "Media," p. 328.
17. Ibid. p. 328.
18. Ibid. p. 336.
19. Ibid. p. 337.
20. Ibid. p. 338.
21. Ibid. p. 338.
22. Ibid. p. 338.
23. Ibid. p. 337.
24. Ibid. p. 327.