Losing Objectivity: Changes in Abstraction Levels of Newspaper Journalism of Surprise Terrorist Attacks on America, 1941 vs. 2001

Michele Rommel

Western Connecticut State University


Using a general semantics perspective, the present study examined the change in journalistic objectivity in reports of surprise terrorist attacks on America, from Pearl Harbor
(1941) to the World Trade Center and Pentagon (2001). A quantitative text analysis utilized an original semantic grammar to code front-page articles of the New York Times on five levels of abstraction for one week following each event. Lower-level abstractions were more abundant in the 1941 press, at a difference of fourteen percent, suggesting decreasing objectivity over time. A twenty-three percent increase in lower-level abstractions from the first to seventh day following the events indicates objectivity is lowest immediately after an attack. The impact of reduced objectivity on news consumers and politics is also discussed.

Losing Objectivity: Changes in Abstraction Levels in Newspaper Journalism of Surprise Attacks on America, 1941 vs. 2001

As regards politics and social trends, the newspaper has become our greatest educator-if you are willing to interpret the word 'education' in its broadest sense and to include the bad with the good. . . Modern life, with its magnificent scope and distances, its subtle, remote first causes, takes its revenge by binding the individual to a narrow sphere of observation. Almost everything he knows about politics, statesmanship, finance, science or any other larger activity, he knows from authority. His main authority is the newspaper (Irwin, 1970, p. 80).

Many studies have been done to prove that newspaper journalism has an enormous effect on the people who read it, both through shaping public opinion and political action (Gunther, 1998; Irwin, 1970; Johnson, 1992; Seaver, 1998). This news medium exerts its influence through little more than the words it prints, making its language the fundamental element in communicating to its audience (Johnson, 1992). Readers rely on the newspaper to provide accurate accounts of current and past events. Unfortunately, news reporting often cannot be based upon verifiable facts and reporters must rely on inference in composing articles (Hoffmann, 1993), thus affecting the neutrality of event coverage.

This study will determine if the level of objectivity in newspaper reporting of surprise terrorist attacks on America has changed from Pearl Harbor of 1941 to the World Trade Center and Pentagon attacks of 2001. It will utilize a general semantics orientation to evaluate the objectivity of news content.

General Semantics

General semantics, advanced by Alfred Korzybski in 1921, remains one of the most widely applicable theories of communication to date (Bois, 1966). It is described as, "applying modern concepts of science to our skills of processing experiences" or "the study of how we perceive, make meaning of, articulate, and communicate our experiences" (emphasis added, Lauer, 1996, Introduction section, para. 1).

General semantics is partially based on the Sapir-Whorf-Korzybski theory of linguistic relativity, which states that each language does not represent the ideas of its users but presents or shapes each user's understanding of the world (Krohn, 1994). Each language, by utilizing particular symbols to represent particular ideas (and not providing other symbols for other experiences) determines its users' thoughts and actions. Expanding upon this theory, general semantics looks to the evaluation of language. Hayakawa provides more tailored descriptions to focus on the particular interests of general semantics: "The central concern of general semantics is to understand symbolism, including language-and to understand how symbolism is both used and abused" (1953, p. v). "[It] pays particular attention not to words in themselves, but to semantic reactions-that is, human responses to symbols, signs, and symbol-systems, including language" (p. 6).


Symbols are representations of thought, emotion, and action that do not have a "direct or natural relationship with what they represent". The ultimate form of symbolism is language. Symbols are characterized as being arbitrary, abstract, and ambiguous; consequently they cannot possibly have definite meanings inherent to them. Hayakawa simplifies this by stating, "the meaning of words (or other symbols) is not in the words, but in our own semantic reactions" (1953, p. 6). Therefore, the evaluation of language is individually determined, while guided by the constructs of and our experience with the language itself.

Abstracting and Abstraction

Abstracting is the process of selecting or highlighting experiences from one's environment. This process is illustrated by Korzybski's Structural Differential (see Appendix E). In this model we see that the procedure begins with our inference of an event (E) and our sensual interaction (O) with that event. We then describe (D) the event using symbols and make inferences, judgments, beliefs, etc. (Ix) about the experience. Over time (A), our inferences become part of future experiences (Stockdale, 2000).

During the descriptive and inferential steps, communicators label their experiences with symbols to attempt to convey the experience to another. As has been shown, though, symbols have no meaning until they are reacted to. Therefore, communicators must choose symbols whose semantic reactions can be agreed upon by other users of the language. These symbols are then said to have a certain degree of verifiability, resulting in a particular level of abstractness, or abstraction. To illustrate the various levels of abstraction, general semanticists use Hayakawa's "ladder of abstraction" (Appendix F). Higher levels of abstraction, or statements that are very general, not easily verified, and vaguely represent many individual things, are placed on the higher rungs of the ladder. Lower level or descriptive statements of particular things are placed further down on the ladder (Bois, 1966). This distinction between low- and high-level abstractions is the basis for this study.

Abstracting in the News

The process of abstracting occurs in every instance of communication. In this study, the process will be applied to the news media. Hoffmann has outlined how the actual events of the day become news stories through the process of abstracting, illustrated on the Structural Differential. He diagrams that the event (E) occurs and is sensed (O) by the reporter, either directly or through reports and interviewing. This is the first instance of abstraction-the language used (D) by the sources of information may occur on a variety of levels of verifiability. Then the reporter condenses the knowledge she has received (I1) into her own set of symbols, raising the communication higher on Hayakawa's Ladder of Abstraction. An editor will then make changes (I2) to the article (again, increasing abstraction) and the reader will interpret the final printed story (I3), which has now become highly abstract in comparison to the event itself (1993).

What is News?

Through [the editor's] selection of local and world news, through the point of view he instilled into the authors of his local news stories, through the kind of news he chose to 'play up' or 'play down' under a single head inside, through his very headlines-in this lay the real power of the press over a modern world (Irwin, 1970, p. 79).

In his book, Propaganda and the News, Irwin outlines two characteristics of "news": "(1) News is any event which varies from the reader's picture of the normal and accustomed world, and (2) news is a report on the conflict of opposing forces" (p. 48).

Irwin goes on to explain that news requires some drama: "Seldom if ever is news static. It means movement, action" (p. 48). In this vein, he identified three mainstays of the news: change, drama, and gossip. He attributes these to the reader's natural curiosity about things surrounding "people whom we know, like, admire, or envy" (p. 50). Irwin also acknowledges the importance of emotion and the proximity of the event or the people involved to the readers, both wielding the power to heighten the reader's interest in a particular story.

Hoffmann (1993) continues to list the elements "news" comprises. First he speaks of timeliness, explaining that news is reporting of the recent past or anticipated future. Celebrity also makes news, he says, be it political or social fame. He mirrors Irwin in including proximity of the event to the reader. Irwin's drama and gossip factors appear again as Hoffmann's "personal" and "dramatic" (1993, p. 24). But Hoffmann also includes the journalist's desire to "scoop" his competitors by writing the piquant new stories. He cites business factors, such as the common "70-30 split" (p. 25) between seventy percent advertising and thirty percent news, the demographic to which the paper will be distributed, the timing of deadlines, and the variety of coverage. He tells of the need to maintain market ratings, popularity, and profit levels. All, he says, are additionally influential in choosing that information which becomes news.

Historical Background of Newspapers

Newspapers have a long history in the United States. In fact, their history in this territory stretches farther back than that of the nation. The first newspaper emerged in Boston in 1690 as a work of printer Benjamin Harris and was called Publick Occurrences. This paper reported what Irwin calls "just news"-military movements and police actions-and abstained from editorializing and opinion pieces (1970, p. 20). More newssheets appeared for brief periods over the next forty years, always running under incredible censorship and libel laws imposed by Britain.

Beginning in the 1730's, as American citizens rebelled against all forms of British oppression, these strict legislations became relaxed, and the next century brought newspapers fraught with opinion and satire. The first widely circulated newspaper whose main focus was the reporting of events was the New York Sun founded in 1833 by Benjamin H. Day. The news industry then began to grow, with men like James Gordon Bennett continually transforming the medium until its current form-a blend of event reporting, opinion, satire, "human" pieces, financial news, etc.--crystallized (Irwin, 1970).

In the nineteenth century, newspapers became subsidized by the government and editors were allowed to trade papers postage-free. Special postage rates were provided for dispersing the newspapers and printing contracts were financed as well. The press had finally become a strong decentralized institution (Johnson, 1992). This relationship between the press and the United States government reaches beyond money, however. As will be shown, each has an incredible influence on the other's actions and on the citizens that constitute its membership.

Effects of News on Politics and Consumers

So much of what we think we know about the world actually is what the media have presented to us (Hoffmann, 1993, p. 6).

Gunther addresses the influence media have over public opinion in his "Persuasive Press Inference" (1998). This hypothesis "suggests that people infer public opinion from their perceptions of the content of media coverage and their assumption of the persuasive impact of that coverage on others." While it may be obvious that people take others' opinions into account when forming their own beliefs, it may not be obvious how or why this occurs.

The persuasive press inference postulates the cyclic nature of this occurrence. First, people use mass media to gather information about the world. These media are, of course, subjectively composed (by editors and staff) toward a certain view, and subjectively processed (by the reader) from a certain view. Readers then assume that others have been exposed to the same media, and thereby the same influences. They project that the media will have substantial influence over others, causing public opinion to be swayed toward the beliefs of the media. Readers then use this projection to formulate their own opinion, which is then supposedly reported by the media. And so the process continues indefinitely.

In her 1998 article, Seaver explains the "flow of influence" that exists between media, public opinion, and political action. Although Seaver cannot ultimately assign causality in this multidimensional relationship, she does identify substantial research supporting correlations of several two-dimensional relationships within it. The media-public opinion association has been widely covered. This article suggests that both the content and feeling of news coverage can greatly affect public opinion of a political issue. Also well supported is the public opinion-foreign policy correlation. One study reveals that most often, "foreign policy outcomes corresponded to public preferences" (The Public Opinion-Foreign Policy Linkage section, para. 8). The media-foreign policy relationship has also been the subject of much research, and Seaver cites an example of this by reminding that policy officials themselves use media as sources of information.

Seaver then combines these correlations to postulate a media-public opinion-foreign policy relationship. She states four critical issues to the extent of this influence: agenda-setting, priming, framing, and global extension. Agenda-setting is the capability by the media to set foreign policy agenda by addressing certain issues more than others, thereby providing different "weights" for the importance of these issues in the public mind. The media prime the public by supplying criteria by which the public will judge policies and policy makers. Framing a foreign policy issue with a particular perspective (or angling the story) will impact the policy makers and the policies, by shading public opinion and reporting a shaded public opinion. And the global extension of American media allows foreign leaders to address the American public and encourage particular foreign policies (1998). These ideas can be applied when determining the multidimensional influence upon foreign policy and all political policies as well.

These relationships, these exchanges of power, between the media, public opinion, and political policy, become especially vital during times of conflict when the activities of each become centered on a single event.

News During Times of Conflict

If everyday news is defined by Irwin as "a report on the conflict of opposing forces" (1970, p. 48), then news during wartime can be described as hyper-news, where the conflict is infinitely larger than any day-to-day discord the reader might encounter. The amplified significance of the news on a reader's everyday life and the country's political and social actions intensifies the production, consideration, and impact of the coverage.

The conflict between America and its enemy obviously becomes the major topic of interest and comprises much of the news during and after the actual battle. Noam Chomsky reminds us that the norm is "for the major media...to line up in support of power at a time of crisis and try to mobilize the population for the same cause" (2001, p. 30). The shift is not surprising, says Hoffmann (1993), because the media are consistent with the sentiments of the market they serve and the officials central to the event. This support is given by using censorship, particular types of speech, and metaphorical descriptions of enemies.

"You can impose censorship for a limited purpose and for a limited time-in war, for instance" (Johnson, 1992). Censorship is one way the media alter public reaction to particularly stressful situations like conflict. Following the September 11th attack, several news articles arose depicting two contradictory national sentiments. In the New York Times, September 28th 2001, a front-page article outlined the tension between American's commitment to freedom of speech and their need for wartime support and patriotism. Several members of the community, including artists and journalists, were sanctioned for seemingly unpatriotic comments following the World Trade Center and Pentagon attacks, despite their right to free speech (Carter & Barringer, 2001).

Language is not only censored during conflict-it is changed. The language used to describe the event, the enemies, and the American efforts are altered to fit the shifting scenario. New, often vague, terms are created to project a framed view of the events. Chomsky (2001) tells that "humanitarian intervention", "crime", "terrorism" and "war" are often interchanged in hostile times, depending upon the intent of the speaker. Stereotypes, paradigms, and metaphors are often used to communicate information, however skewed, about the conflict as well (Hoffmann, 1993). The adjustments appear to habitually favor higher-level abstraction.

"In reading the news, the average mind writes its own novel as it goes along; and always with a central character, be he hero or villain. This was one of the first lessons which editors learned about the news" (Irwin, 1970, p. 55). MacDougall realizes the importance of creating a character of the enemy as he describes the "red, brown, and yellow perils" (1999, p. 59), America's foes from World War II to the end of the Cold War, as depicted in popular media. The entire Japanese culture was considered our enemy during World War II, and was portrayed as a diminutive, buck-toothed subhuman, often compared with apes. Later, the Nazi Germans become our adversaries; the media represented this regime with its leader, Adolf Hitler, or another diabolical figure. MacDougall notes, however, that these images were used to stir up more than pro-war sentiments: the enemies represented fears and discord within the United States during these times of international tension. Therefore, by utilizing such abstract descriptions, the media both framed the issue in the readers' minds and made the conflict personal and dramatic (1999).

Pearl Harbor vs. the World Trade Center and Pentagon

The two conflicts at the focus of this study are Pearl Harbor and the recent attacks on the World Trade Center and Pentagon. The event now referred to as Pearl Harbor occurred on December 7, 1941, when Japanese planes suddenly began dropping bombs on the naval base at Pearl Harbor, Hawaii. The attack stimulated World War II. On September 11, 2001, both World Trade Center towers and the Pentagon were hit by jet airplanes under the control of terrorists. These assaults sparked the current efforts against the Taliban and other terrorist organizations. Both are considered "surprise attacks" on America, involving acts of terrorism. The U.S. Congress defines terrorism as...

"any activity that (A) involves a violent act or an act dangerous to human life that is a violation of the criminal laws of the United States or any State, or that would be a criminal violation if committed within the jurisdiction of the United States or of any State; and (B) appears to be intended (i) to intimidate or coerce a civilian populations; (ii) to influence the policy of a government by intimidation or coercion; or (iii) to affect the conduct of a government by assassination or kidnapping" (Chomsky, 2001, p. 16).

Many suggest, however, that the comparison between these two events is misleading. Chomsky upholds this view by clarifying that Hawaii was not a territory of the U.S. at the time of Pearl Harbor, but a colony, so should not be included in the former category (2001). An article from Naval History compares the September 11th, 2001 attacks to the 1941 attack. "We are having another Pearl Harbor. But not quite," writes Schultz (2001):

Yes, both attacks were devastating surprises. The Pearl Harbor attack, however, was a tactically brilliant military operation, on a military target, by military men with a certain sense of honor and bravery. The 11 September attack, on the contrary, was neither military nor honorable nor brave (2001, para. 5).

Nevertheless, both events involved U.S. citizens (whether military or civilian), both meet the criteria of terrorism as provided by the U.S. Congress, and both are widely deemed surprise attacks because of their publicly undisclosed preparations. Joe Strupp (2001) even provides that the increased press runs of September 11, 2001 mimicked those seen on December 7, 1941.

The New York Times

The New York Times, a newspaper established in 1851, covered both of these events extensively. In fact, the Times won six Pulitzer Prizes for its journalistic treatment of the attacks on the World Trade Center and Pentagon (Barringer, 2002).

The New York Times is heralded as being one of the top newspapers in the country (Irwin, 1970). Its motto, "All the News That's Fit to Print" suggests that the Times is comprehensive in its reporting and careful in choosing what news to convey. The Times, in fact, follows an extensive style guide which first appeared in 1895 to ensure consistent reporting. This guide asserts that it "favors constructions that keep language neutral, a crystalline medium through which journalists report ideas without proclaiming stances" (Siegal and Connolly, 1999, p. viii). But this respected news source may not attain that goal when reporting events of American conflict. To determine how well the Times achieves this ambition, this study utilized text analysis to investigate the level of objectivity used in the Times' coverage of surprise terrorist attacks on America.


Using a method of quantitative text analysis, the levels of abstraction in the New York Times' coverage of the Pearl Harbor and World Trade Center and Pentagon attacks are compared to determine if the journalism surrounding American conflict is gaining or losing objectivity.

Text analysis, as defined by Shapiro and Markoff is a scientific method of measuring text for statistical manipulation in a social science study (Roberts, 1997). In order to do this, a tool must be developed to ensure consistency in the coding of the text. This tool is the semantic grammar. Roberts identifies a semantic grammar as "a template that the researcher uses to capture essential elements of the phenomenon under investigation" (1997, p. 4). For this study, the semantic grammar was based upon Hayakawa's "Ladder of Abstraction" (1964) (see Appendix E) and text analysis procedures provided in Roberts' chapter entitled "Semantic Text Analysis: On the Structure of Linguistic Ambiguity in Ordinary Discourse" (1997).


The researcher analyzed the front-page articles of the New York Times New York Metro Late editions for one week after each attack, beginning on the date following the date of the event (December 8th through 14th, 1941 and September 12th through 18th, 2001). Phrases in the text that spoke directly of or speculated upon the event, its causes or outcomes, key figures involved, public or political responses, etc. were included in the study, while reports of other news items were excluded. Headlines, photo captions, questions posed by the reporter, and other ancillary materials were also excluded.

Only noun-verb-modifier phrases were considered; components were permitted to occur in any order. Both action and being verbs were eligible, however verbs without subjects used as modifiers (i.e. "As the young boy, found in a crowd of 200, cried to his mother...") were not deemed verbs. A noun was allowed to be counted for multiple verbs if part of a compound sentence.


The researcher developed a semantic grammar for coding each noun-verb-modifier phrase for a level of abstraction (see table below). Lower-level abstractions are comprised of Levels 1, 2, and 3 while higher-level abstractions consist of Levels 4 and 5.

Level of
Identifying Characteristics Example
1 Statement of action/being "The bomb was dropped. . . ", "The president spoke. . . "
2 Report, speaker identified "The President said the troops rushed. . . "
3 Report, speaker anonymous "Sources say the troops were ambushed. . . "
4 Evaluation of action/being "The life-snatching bomb. . . ", "An inspiring speech ended. . . "
5 Statement of opinion "This horror should not be endured. . . "

After reading and scrutinizing each phrase, the researcher noted its level on a tally sheet. When all eligible phrases were derived from one day of one event, the total number of phrases in each abstraction level was found and taken as a percent of the overall number of phrases for that date of the event. Finally, each level for the seven days following an event was summed and taken as a percent of the total number of phrases for the event. Percents were rounded to the nearest integer.


The results are located in Appendix A. Total results for each event are displayed in Appendix B. Results by day following the event are presented graphically in Appendices C and D. Note that these visuals represent the percents given in Appendix A.

The total percents of abstractions by level for each event reveal that the Pearl Harbor event coverage consistently involved more lower-level abstractions (Levels 1, 2, and 3) than that of the World Trade Center and Pentagon, which produced more higher-level abstractions (Levels 4 and 5). It was found that the coverage of Pearl Harbor yielded higher levels of Level 1 abstractions five of seven times and higher levels of Level 2 and 3 abstractions four of seven times, while the coverage of the World Trade Center and Pentagon bore greater Level 4 and Level 5 abstractions all seven of seven times. The total percents differed between events by an average 5.6% variation (Appendix B).

The total results per day following the event (both events combined) showed a 23% increase in lower-level abstractions used on Day Seven, as compared to Day One, with Level 1 increasing by 33%, Level 2 decreasing by 23%, and Level 3 increasing by 13%. Higher-level abstractions decreased between Days One and Seven following the event, with Level 4 abstractions decreasing by 10% and Level 5 abstractions decreasing by 13% (Appendix C).


The results of this study suggest that the news media, as represented by the New York Times, are losing objectivity in their reporting of surprise terrorist attacks on America. Although only these two special situations were included in this analysis, it may, with further investigation, be possible to extend the results to journalism on a larger scale.

The total percent of lower-level abstractions was greater, averaging 87%, for the 1941 coverage of Pearl Harbor, while in the reporting of the World Trade Center and Pentagon attacks, the percent of lower-level abstractions averaged only 73% of the news reported. This 14% decrease may be attributed to many factors. Among them are the following three explanations:

First, the advent of other news media, such as television and the Internet, has almost certainly changed the role of the modern newspaper in everyday life in two ways. To begin with, more "real-time" news allows people to watch an event as it happens or receive up-to-the-minute updates on breaking news stories. Therefore, media consumers often do not need to rely on the detailed accounts of the prior day's news events provided by a newspaper to understand current events. This is particularly true in times of conflict when media coverage is heightened. Also, electronic media, especially television, involve many more means of communicating information than mere words, such as pictures, sound, and motion. These images prescribe the frame in which the story will be told (Johnson, 1992). Media consumers' tastes for news, then, may have begun to crave the sensory stimulation these electronic media provide as well as the abbreviated time in which messages are given. This leaves newspapers, which depend primarily on silent, motionless words, to compete for the attention they are steadily losing (Hoffmann, 1993). Perhaps by being more subjective (it may be speculated, more like the electronic media) newspapers can retain readership.

Second, the goals of the newspaper may now be different than they were in 1941. The Times may be more concerned with popularity or business factors, as discussed by Hoffmann, or winning Pulitzer Prizes, thus allowing stylistic writing, metaphorical phrasing, and more personal involvement by the reporter. The level of objectivity sought by the New York Times Manual of Style and Usage (1999), as discussed earlier, may have been abandoned in favor of other needs or goals.

Third, the location of the events with regards to readers and the newspaper itself could have affected the results of this study. Both Irwin (1970) and Hoffmann (1993) discussed the importance of proximity in constructing news. The New York Times New York Metro edition may have covered the attacks on the World Trade Center more subjectively because of the closeness of the newspaper's headquarters and its readers to that disaster. Both factual information and more "human" stories of loss were available quickly and abundantly. The newspaper's typical reader almost certainly knew at least one person who escaped, was injured or killed, assisted in rescue efforts, etc. simply because of locale. However, Pearl Harbor is thousands of miles from the New York Metro area, and with the cruder technology of the 1940's, information and stories were not easily relayed back to the homeland. The people immediately affected were military personnel or their families, far removed from the majority of the Times New York Metro. Pearl Harbor was just not as close, physically or emotionally, to the New York edition of the Times and its readers. This distance most likely increased the objectivity of the 1941 reporting.

The results also show a 19% increase in the amount of lower-level abstractions from Day One following the events (both events combined) to Day Two, and a 23% increase from Day One to Day Seven. This finding indicates that the initial feelings of "shock, horror, anger, fear, [and] a desire for revenge" (Chomsky, 2001, p. 20) following the event are supported by the reporters and editors of the news.

But by reporting so much subjective emotion, it has been discussed, the news media are not simply exposing popular sentiments-they are providing a construct in which to think and feel about the events. In this particular case of surprise terrorist attacks, when the public has probably not even fathomed that such an event could occur, the framework set by immediate news coverage will most likely be quickly adopted by the news consumers. It will become the mode through which they will think and act in reaction to the incident. This idea returns us to the general semantics concept of semantic reactions.


Semantic reactions ("human responses to symbols, signs, and symbol-systems" [Hayakawa, 1953, p. 6]) are at the core of human existence. They determine, according to general semantics, how we interpret our language. Since the theory proposes that we can only understand that which is represented in our language, semantic reactions essentially determine how we interpret our world (Krohn, 1994). The process of reporting the news is itself quite abstract (Hoffmann, 1993). To add to this already high level of abstraction, the media use several tools to shift the audience's semantic reactions, both in and out of times of conflict. Among these are agenda-setting, priming, and framing (Seaver, 1998); censorship; changing the language itself; and vilifying the enemy (MacDougall, 1999). These tools require the newspapers to use higher-level abstractions to present the article's slant or shade. They force editors to hide facts in subtleties to avoid any forbidden reports. And they entice reporters to create metaphorical comparisons that support the government's efforts. In short, they compel news media to become increasingly subjective.

As early as 1807, Thomas Jefferson recognized the impact of subjective journalism: "The man who never looks into a newspaper is better informed than he who reads them; inasmuch as he who knows nothing is nearer to the truth than he whose mind is filled with falsehoods and errors" (Johnson, 1992, Role in a Republic section, para. 3). While this seems to be somewhat of an overstatement, the core of Jefferson's assertion remains true. The newspaper, the news medium that is perhaps most dependent upon language, is using that one great asset, the power of its words, to change how people interpret their world and the way they live their lives. And the interpretation is becoming based more in opinion and evaluation-in abstract, subjective language-than in verifiable facts. We are living farther from the truth.

Other Findings

The researcher noted several differences between the newspapers of Pearl Harbor and those of the 2001 World Trade Center and Pentagon attacks that were beyond the scope of this study. The size of the New York Times itself has changed dramatically. The Times of the 1940's contained eight columns of approximately 100 lines per column while in the 2000's contained only six columns. Where the average number of phrases coded in the 1941 New York Times was 212, the 2001 newspapers averaged a mere 150. Percents were used for calculating the results of this study to avoid bias due to physical size.

The size and subject of photos also differs between these events. Pictures on the front pages of the 1941 newspapers were smaller and depicted maps of the territories involved, people (such as President Roosevelt), or objects (a Christmas card or The Normandie). Photos in the 2001 issues of the Times added "action" shots of buildings burning, rescue workers sorting rubble, candlelight vigils, etc. The photos of people in 2001 included the President, but more often were of unnamed people involved in the activities surrounding the event. The 2001 newspapers also used more pictures (an average of 3 per paper) than the 1941 papers (an average of 1.8 per paper).

These changes suggest that the Times, in altering its format, has fostered subjectivity, including less written news and more suggestion through photos.


To read daily every word in the New York Times would be a career in itself. Probably no one does. Perhaps a majority skim most of the front page. Then the reader runs over the headlines... The headline has served its purpose. It has enabled him to select from the offerings of the newspaper the fare he really wants; and it has given him a hazy idea of events beyond his own circle of immediate interest (Irwin, 1970, pp. 59-60).

Several limitations were present in this study. First, the inclusion of headlines, photos and captions, questions posed by the reporter, and other text auxiliary to the articles studied would have altered the results. As headlines often determine which article a reader will examine (Irwin, 1970), these titles impact which phrases, with corresponding abstraction levels, will be read. Photos and their captions also contain levels of abstraction. Not only are the words in the caption low- or high-level, but, on another plane, the subjects of the photos also represent concrete and abstract ideas. For example, a picture of a fallen building may be "low-level" in its depiction of the event of the building collapsing, and "high-level" in its suggestion of defeat and despair. The questions posed by the reporter suggest questions that the reader should herself be asking, thereby prescribing to the reader a "normal" mode of thinking about the report. All these supplements, while beyond the range of this study, influence the reader's semantic reactions to the events. Were they included, the abstraction levels of each event and day would have been different.

Second, while opinions of reports were categorized as Level 4, reports of opinions were labeled Level 2 (speaker identified) or Level 3 (speaker anonymous), and therefore were counted in the lower-level abstraction categories. While this is correct for the semantic grammar prescribed in this study, these expressions of opinion are subjective and could be coded differently in another semantic grammar. In addition, these second-hand opinions may affect the newspaper reader's own opinion. As stated in Gunther's "Persuasive Press Inference" (1998), people often form their opinions based upon the "public opinion", which is reported in the newspapers and other media.

Third, actions said to occur in the future (i.e. "The President will..." or "Troops should...") were coded under the appropriate abstraction level, disregarding their tense. General semanticists, however, would argue that as the future cannot be verified, these predictions should not be coded, or should perhaps be coded on an "opinion" level. Phrases with futuristic verbs were included in this historical study because of their ability to be verified presently.

Lastly, in instances of poor newspaper reproduction quality, a small number of phrases were ultimately excluded. While the researcher believes the number of phrases to be insignificant to this study's results, this omission does affect the outcome of the research.

If the project were to be done again, inclusions of material ancillary to the article text could be included for a more thorough look at the change in objectivity of the entire front page. Another semantic grammar should also be developed to better represent the concepts of general semantics.


The analysis completed set out to determine how the level of objectivity used in news media has changed since 1941. While containing several limitations, the study found that newspaper coverage in the New York Times of surprise terrorist attacks on America has become more subjective, defined as using more higher-level abstractions, in the past sixty years. The implications of this change upon journalism as a whole were addressed. To a greater extent, the influence of news media upon politics and the public has been discussed, as well as the ramifications of more subjective media on consumers' daily interpretations of the world. However, what remains to be shown is the extent to which this effect occurs. It would be beneficial to discover the amount and type of influence the articles of the front page have on the average reader's semantic reactions regarding a surprise attack on America, or in a larger application, regarding their entire world. By determining the level of impact, that discovery could awaken newspaper staffs to the power they wield and consumers to the importance of critical absorption of news reports. In general semantics terms, increasing the objectivity of the news would create greater agreement among media and consumers and promote truer semantic reactions. This would improve communication, thereby improving our lives.


Barringer, F. (2002, April 9). Pulitzers focus on Sept. 11, and the Times wins 7. New York Times. Retrieved April 10, 2002, from http://www.nytimes.com/2002/04/09/nyregion/09PULI.html

Bois, J.S. (1966). The art of awareness: A textbook on general semantics. Dubuque: Wm. C. Brown.

Carter, B. and F. Barringer (2001, September 28). In patriotic time, dissent is muted. New York Times, 151, A1.

Chomsky, N. (2001). 9-11. New York: Seven Stories.

Gunther, A.C. (1998, October). The persuasive press inference [Electronic version]. Communication Research, 25, 486-505.

Hayakawa, S.I. (1953). Symbol, status, and personality. New York: Harcourt, Brace, & World.

Hayakawa, S.I. (1964). Language in thought and action. New York: Harcourt, Brace, & World.

Hoffmann, G. (1993). Media maps & myths. Whitefish Bay, WI: M&T Communications.

Irwin, W. (1970). Propaganda and the news or what makes you think so? Westport, CT: Greenwood.

Johnson, P. (1992, December). The media and truth: Is there a moral duty [Electronic version]? Current, 4-8.

Krohn, F.B. (1994). Improving business ethics with the Sapir-Whorf-Korzybski hypothesis in business communication classes [Electronic version]. Journal of Education for Business, 69, 354-359.

Lauer, R.M. (1996). Some basic ideas of general semantics [Electronic version]. ETC: A Review of General Semantics, 53, 292-300.

MacDougall, R. (1999). Red, brown, and yellow perils: Images of the American enemy in the 1940s and 1950s. Journal of Popular Culture, 32, 59-75.

Roberts, C.W. (Ed.). (1997). Test analysis for the social sciences: Methods for drawing statistical inferences from texts & transcripts. Mahwah: Lawrence Erlbaum Associates.

Schultz, F.L. (2001). On our scope [Electronic version]. Naval History, 15, 6.

Seaver, B.M. (1998). The public dimension of foreign policy [Electronic version]. Harvard International Journal of Press/Politics, 3, 65-91.

Siegal A.M. and W.G. Connolly (1999). The New York Times manual of style and usage. New York: Random House.

Stockdale, S. (2000). Korzybski's structural differential and Hayakawa's abstraction ladder. Retrieved April 21, 2002 from The Dallas Fort-Worth Center for General Semantics Web site: http://www.thisisnotthat.com/gs/abs_mod.html.

Strupp, J. (2001). US attacks get 'extra' attention [Electronic version]. Editor & Publisher, 134, 9.

Appendix A: Results Tables

TOTALS Pearl Harbor World Trade Center
Level 1 Abstractions 738 50% 487 46%
Level 2 Abstractions 343 23% 173 17%
Level 3 Abstractions 201 14% 104 10%
Level 4 Abstractions 167 11% 179 17%
Level 5 Abstractions 32 2% 109 10%
Day One Pearl Harbor
World Trade Center
Level 1 Abstractions 84 47% 32 33%
Level 2 Abstractions 47 26% 20 21%
Level 3 Abstractions 18 10% 3 3%
Level 4 Abstractions 26 15% 23 24%
Level 5 Abstractions 4 2% 19 19%
Day Two Pearl Harbor
World Trade Center
Level 1 Abstractions 104 50% 76 41%
Level 2 Abstractions 65 31% 29 15%
Level 3 Abstractions 13 6% 30 16%
Level 4 Abstractions 23 11% 28 15%
Level 5 Abstractions 4 2% 24 13%
Day Three Pearl Harbor
World Trade Center
Level 1 Abstractions 94 35% 70 47%
Level 2 Abstractions 110 42% 21 14%
Level 3 Abstractions 30 11% 23 16%
Level 4 Abstractions 23 9% 24 16%
Level 5 Abstractions 7 3% 11 7%
Day Four Pearl Harbor
World Trade Center
Level 1 Abstractions 127 51% 77 47%
Level 2 Abstractions 46 18% 28 17%
Level 3 Abstractions 45 18% 12 7%
Level 4 Abstractions 30 12% 28 17%
Level 5 Abstractions 3 1% 20 12%
Day Five Pearl Harbor
World Trade Center
Level 1 Abstractions 137 62% 81 54%
Level 2 Abstractions 38 17% 25 17%
Level 3 Abstractions 15 7% 11 7%
Level 4 Abstractions 23 10% 22 15%
Level 5 Abstractions 8 4% 11 7%
Day Six Pearl Harbor
World Trade Center
Level 1 Abstractions 96 53% 60 41%
Level 2 Abstractions 18 10% 29 20%
Level 3 Abstractions 45 25% 16 11%
Level 4 Abstractions 20 11% 27 18%
Level 5 Abstractions 3 1% 14 10%
Day Seven Pearl Harbor
World Trade Center
Level 1 Abstractions 96 55% 91 58%
Level 2 Abstractions 19 11% 21 13%
Level 3 Abstractions 35 20% 9 6%
Level 4 Abstractions 22 12% 27 17%
Level 5 Abstractions 3 2% 10 6%

Appendix B: Total Results by Event

TOTALS Pearl Harbor World Trade Center
Level 1 50% 46%
Level 2 23% 17%
Level 3 14% 10%
Level 4 11% 17%
Level 5 2% 10%

Appendix C: Total Results by Day

Day One Day Two Day Three Day Four Day Five Day Six Day Seven
Level 1 80% 91% 82% 98% 116% 94% 113%
Level 2 47% 46% 56% 35% 34% 30% 24%
Level 3 13% 22% 27% 25% 14% 36% 26%
Level 4 39% 26% 25% 29% 25% 29% 29%
Level 5 21% 15% 10% 13% 11% 11% 8%

Appendix D: Daily Results Charts

Appendix E: Korzybski's Structural Differential (Stockdale, 2000)

Appendix F: Hayakawa's Abstraction Ladder (Hayakawa, 1964)

Table of Contents