The use of Internet in newsgathering among European science journalists
21
it is the organization of news, not events in the world, that creates
news. Organization includes not only that of the news-media
institution and particular news operations, but also sources and their
apparatus for making news. (Ericson, Baranek et al. 1987, p. 345)
The studies on the shaping of news can be traced to the works of White (1950).
By analysing the choices (and the reasons for those choices) of the wire editor of a
non-metropolitan newspaper in a highly industrialized city of the Midwest of the
United States, he applied the term ‘gatekeeper’ (coined by social psychologist
Kurt Lewin) to journalism. Later, Gieber (1964)5 followed the steps of White and,
by studying wire editors at 16 daily newspapers in Wisconsin, infered that they
showed almost no critical evaluation of the incoming news, and that they had no
real perception of their audiences.
Although the studies of White and Giber are essential steps in the understanding
of the construction of news process, there are mainly two problems in the
gatekeeper metaphor: the first, is that it reduces the decision of publishing or not
publishing to a lone individual, as if the organization did not matter; and the
second is that it does not describe other external processes involved in the
production of news, such as readers’ feedback or the anticipation of the
gatekeeper’s criteria by news sources (Schudson 1989).
A second series of studies on the construction of news begins in the 1970s with
the agenda-setting research. In the article that inaugurates this field of research,
McCombs and Shaw (1972, p.184) show that ‘voters tend to share the media’s
composite definition of what is important’. According to this study, and all the
others that follow this hypothesis, there is a correlation between the importance
given to certain issues in the media to the importance given to the same issues by
the public or politicians. In a sense, agenda-setting research tries to prove that the
media agenda influences the public and the politicians’ agenda.
5
The work of Gieber was mainly done in his PhD thesis, presented in 1956 at the University of
Wisconsin. Nevertheless, his work of 1964 is the one most frequently cited by scholars analysing
gatekeeping studies.
22
Special interest groups – like scientists or scientific institutions, the main sources
of information for the journalists we are studying in this thesis – can also have a
big influence on the media agenda. In her study, Kyle Huckins (1999, p.83) found
‘strong correlations among three-month lagged 1992 and 1994 Christian coalition
agendas and the subsequent media agendas’. So, she says, this provided evidence
how ‘an interest group can make a purposeful impact on media coverage.’
Studies like the one carried out by Huckings play against the agenda-setting
hypothesis, because they turn around the agenda-setting model by saying that it is
the public’s agenda that influences the agenda of the media, and not the other way
round. This is exactly one of the most important criticisms that has been made to
the hypothesis of agenda-setting, as it is very difficult to prove that the correlation
between agendas means causation in any direction.
The further one moves from the general notion that the media direct
attention and shape cognitions and towards examining actual cases,
the more uncertain it becomes whether such an effect actually
occurs. (McQuail 2000, p.455)
With the end of the 1970s comes the consolidation of another approach to the
analysis of the construction of reality by the media. The titles of the main works in
this period show that the sociologists involved in them follow the footsteps of
social constructionism: Making News: A study in the construction of reality, by
Gaye Tuchman (1978); Deciding what's news: A study of CBS Evening News,
NBC Nightly News, Newsweek and Time, by Herbert J. Gans (1979); and
Manufacturing the News, by Mark Fishman (1980) are among the most cited
books for this period.
According to these authors, journalism is a regulated and routine process of
manufacturing a cultural product. The planning, gathering, selection, writing and
editing stages of a news outlet is done inside an organization within a given time.
‘The news is made, and like any other product, it carries the marks of the
technical and organizational structure from which it emerges’. (Golding and
Elliott cited by Manning 2001, p.51).
23
Participant observation is common to Tuchman, Gans and Fishman studies. The
researchers involved spent days inside the organizations trying to understand how
journalists and editors worked, what their values were, what they were concerned
with. The bureaucratic structure of the organization emerged: ‘A newspaper is a
bureaucracy’, wrote Tuchman (1978, p.149).
Journalists do not see themselves as bureaucrats. In fact, they mock bureaucracy
and sometimes become frantic about it. But the truth is that there are five basic
characteristics of a bureaucracy, all present in news organizations: a staff; fixed
areas of jurisdiction; a hierarchy; a rational system of expertise; and an ethos of
objectivity – which means that each case is supposed to be handled according to
the applicable procedures. (Berger and Berger 1976, p.215-217)
When analysing news organizations from this angle, researchers became aware of
many of the constraints that influence the work of journalists: sources, editors,
colleagues, other media, audiences, time, space, money, their own background
and attitudes. The power of sources in shaping the news, in a competitive
environment that resembles a production line, became more evident. The
establishing of facts by the simple use of credible sources or conflicting positions,
without the need of independent verification, became understandable. The search
for consensus of what is news, even between competing news organizations,
became logical.
In the end of the 1980s, an important study brought a new perspective to the
research on the shaping of news. Stocking and Gross (1989) explained how the
cognitive processes of journalists themselves could influence the outcome of their
work. According to their investigation, the cognitive process of a journalist could
be divided in six different stages: stimulus; categorization; theory generation;
theory testing; selection of information; and integration of information.
The main idea behind their thesis is that people tend to seek and select
information in ways that confirm their initial beliefs. So, when confronted with a
stimulus, people (and also journalists) tend to immediately categorize it. For
24
instance, when it was known (in 2003) that the wife of the British prime minister,
Tony Blair, was involved with a lawbreaker, this fact was categorized by
journalists as ‘Cheriegate’, from Watergate, the name by which became known
the press investigation that led to the resignation of Richard Nixon, in 1973.
Immediately after, a hypothesis was generated (‘if Cherie Blair is involved with a
crook, the prime minister must be too’), and reporters started testing that theory.
Similarly to scientists, the researchers say, ‘journalists regularly test hypotheses’
(Stocking and Gross 1989, p.19). The selection of information is then done in
order to test this theory, and sources are selected in order to confirm the assertion.
The scope of questions is limited, sources are handled in order to get
confirmations. Sources who disconfirm the ideas of the journalist are even
regarded as unreliable, as research had already shown:
A reporter who had been researching a feature story on crimes
against the elderly told me he had found that police figures showed a
decrease in these crimes compared with the previous year. The
reporter was puzzled and eventually decided to ignore the police
figures. He felt they were unreliable and incomplete, and anyway he
had to do the story as originally planned because the whole issue was
too big to pass up or play down. (Fishman 1980, p.5)
Finally, the information is integrated. Journalists construct links between events
that sometimes have nothing to do with each other and issues are oversimplified.
The result of this process is also a construction of reality, but in this case it is a
biased reconstruction by the preconceptions of the journalist. In science news, like
in other fields of journalism, this biased reconstruction of events also happens,
and has been studied, as we will have the opportunity to see.
25
2.1. Studies on the constraints of journalists
To probe deeper into the construction of science news, we will have to examine
the research already done on the constraints journalists face while doing their job.
For practical reasons, we will first analyse what previous research has said about
the pressures science journalists receive from sources. Subsequently, we will
present the studies about the influences of the news organization, the competitors,
audiences and even the personal background of journalists in the construction of
science news.
The peer-reviewed studies carried out on the Internet and its influence on the daily
routine of journalists will be analysed afterwards. Throughout these chapters
special attention will be given to studies about the work of science journalists and
the special characteristics of their job, as the main reason for this literature review
is to set the stage for the original research project that will be part of this thesis.
2.1.1. Sources’ influence on the news
It is a common saying that without news sources there is no news. However,
research has shown that sources are not only necessary in the news process – they
have an enormous influence on it (Gans 1979; Fishman 1980; Ericson, Baranek et
al. 1989):
The relationship between sources and journalists resembles a dance,
for sources seek access to journalists, and journalists seek access to
sources. Although it takes two to tango, either sources or journalist
can lead, but more often than not, sources do the leading. (Gans
1979, p.116)
26
The ‘leading’ of the sources is well documented. In his book Reporters and
Officials: The organization and politics of newsmaking, Sigal (1973) showed that
58.2 percent of the page one stories in The New York Times and The Washington
Post, during two weeks at five-year intervals (1949, 1954, 1959, 1964 and 1969)
where coming from routine channels. Following up Sigal’s work, Jane Delano
Brown and her colleagues analysed front page stories in The New York Times, The
Washington Post and four North Carolina newspapers in 1979 and 1980. They
also found that ‘most reporting relies on routine channels and press releases’.
(Brown, Bybee et al. 1987, p.53)
Journalists do not turn to routine channels by chance. They live within an
organization that has to manufacture a product to a tight daily (sometimes hourly)
deadline. Journalists just do not have the time, or the resources, to start looking
for news independently, ignoring all the information initiated by the sources. So,
says Gandy (1982), the journalist usually turns to information subsidies6. ‘Faced
with time constraints, and the need to produce stories that will win publication,
journalists will attend to, and make use of, subsidized information’. (Gandy 1982,
p.62)
In science journalism, the reliance on routine channels and information subsidies
seems to be bigger than in other areas of journalism (Goodell 1977, p.128), and it
has even evolved to what Weingart (1998) calls the ‘science-media-coupling’. His
thesis is simple:
With the growing importance of the media in shaping public opinion,
conscience and perception in one hand and a growing dependence of
science on scarce resources and thus on public acceptance on the
other, science will become increasingly media-oriented. (Weingart
1998, p.872)
6
‘An information subsidy is an attempt to produce influence over the actions of others by
controlling their access to and use of information relevant to those actions. (…) This information is
a subsidy because the source of that information causes it to be made available at something less
than the cost a user would face in the absence of the subsidy.’ Gandy, O. H. (1982). Beyond
agenda setting : information subsidies and public policy. Norwood, N.J., Ablex Pub. Co., p.61.
27
This close relationship between science and the media, and between scientists and
journalists, is something worth analysing more deeply. In his work, Sigal (1973)
already told us that the excessive contact with sources could lead to the absorption
of the sources’ values and, ultimately, to what he called ‘beat parochialism’:
Many of the journalists covering Apollo were so caught up in the
drama of the space race that they came to regard themselves as part
of the NASA “team”. (Sigal 1973, p.48-49)
Several reasons have been given to explain this special ‘coupling’ of scientists and
journalists. Rae Goodell points out three main reasons, that have to do with the
complexity of the subject journalists are writing about: (1) science is intimidating
and a close relationship with the scientists might help to overcome that; (2)
journalists have a lack of confidence in themselves to judge what is really
important; (3) trust may be jeopardized by bad coverage (Goodell 1979).
Weingart (1998, p.878) says that the reason for this coupling can be ‘understood
as an expected side phenomenon of modern mass democracies and corresponds to
their increased demands of legitimacy’. Other authors have pointed out that
science seems to be above other legitimation processes in society. Its apparent
neutrality, moved by universal criteria, makes scientists more respected sources,
seldom confronted by journalists (Goodell 1977; Nelkin 1987). The fact that most
science news items are about medical discoveries, and praise the positive aspects
of them, also helps this uncritical relation between science journalists and their
sources.
Even when dealing with scientific fraud, journalists tend to ‘appear relatively
tame’ (Lafollette 1992, p.153), or use words they would not use in any other
fields:
Fraud is a “sin” as well as a scandal. The culprit has “fallen” or
betrayed the profession […] Consumer fraud is a “ripoff” or a
“crime”, hardly a “sin.” Political scandals are abuses of trust and
reported, often cynically, as critiques of political institutions. (Nelkin
1987, p.26)
28
John Crewdson, who investigated Robert Gallo’s claim to have discovered the
AIDS virus, and uncovered a fraud in a breast cancer study, says that science
reporters do not like to criticize scientists because they believe in the good nature
of science. When science journalists know of a good science discovery made by a
scientist, ‘they don’t interview his enemies or call his lab technicians at home (…)
They like science, they probably admire [the scientist] and they’re excited by the
prospect that he’s right’. (Crewdson cited in Dunwoody 1999, p.65)
Several authors talk about a shared culture between scientists and journalists,
which becomes more evident in episodes of scientific controversy (Lewenstein
1995). For instance, in 1998, science writers responded very negatively to an
article in The New York Times that contained enthusiastic endorsements of an
experimental cancer drug that was only being applied in mice (Cooper and
Yukimura 2002). The angry reactions of both the scientific and journalistic
communities in this episode, which raised false hope for cancer sufferers, supports
the assumption that one characteristic of the shared culture is that, when an
individual violates one of the culture’s rules, both journalists and sources turn on
to that person. (Blumler and Gurevitch cited in Dunwoody 1999)
The shared culture between scientists and journalists was also demonstrated in
some research conducted in the 1960s and the 1970s. In the first case, a group of
editors, science journalists and scientists had to rate different science stories using
25 semantic differentiated scales, with pairs like valuable-worthless, or
responsible-irresponsible. Editors tended to evaluate science news on
sensationalistic dimensions. Scientists and science journalists were more
interested on accuracy and significance. (Johnson 1963)
In 1979, Ryan tested the attitudes of scientists and journalists towards media
coverage of science news. In 32 of 38 issues, scientists and science writers were
on the same side of the agree/disagree scale or the differences were not
statistically significant. (Ryan 1979)
29
Main sources of science journalists
The main source of information for science journalists are the scientific articles
published in peer-reviewed journals (Jones, Connell et al. 1978; Hansen 1994; van
Trigt, de Jong-van den Berg et al. 1994; Entwitsle 1995; Hotz 2002; Weitkamp
2003). Most of the time, these scientific journals issue press releases, calling the
attention of reporters to the articles in the upcoming editions. According to
research done on this subject, journal articles described in press releases are more
likely to generate newspaper stories on the same topic. (de Semir, Ribas et al.
1998; Bartlett, Sterne et al. 2002; Stryker 2002)
A scientific article can be considered, using Fishman’s classification, as a
‘performative document’. This means that what the document says must be treated
as a fact in itself. ‘Journalists love performative documents because these are the
hardest facts they can get their hands on’ (Fishman 1980, p.99). Science
journalists may feel manipulated by these press releases, but they admit their
dependence on pre-selected news, coming from scientific publications (Nelkin
1987).
Journal editors know this. They have long understood the way journalists work
and the need they have for reliable information on science. And they gladly
provide that information. Scientific journals are also competing with each other –
‘Today, a key task of editors at the prestige journals is travel to conferences and
labs for recruiting star scientists to write for their publication’ (Jaffe 2002) – and
they know how important it is to get media coverage.
But science journals do not simply provide that information to the media. They
must try to get their research in as many media outlets as possible, and so they
have to be sure that every journalist gets the information at the same time. If there
is a leak, the news will probably be published in only one or two places, as
journalists love scoops. So, scientific journals try to control the information that
30
gets to the press in two different ways: by applying the Ingelfinger rule and
forcing embargoes.
Defined by Franz J. Ingelfinger, former editor of The New England Journal of
Medicine (NEJM), the Ingelfinger rule is an agreement between the journals and
the scientists7. It was published more than 30 years ago and it says:
Papers are submitted to the Journal with the understanding that they
or their essential substance, have been neither published nor
submitted elsewhere (including news media and controlledcirculation
newspapers). (Definition of sole contribution, 1969,
p.676)
In 1981, A.S. Relman, who followed Ingelfinger as the editor of NEJM, explained
that the main reasons for this rule were the protection of the Journal’s
‘newsworthiness’ and the need for medical research to be peer-reviewed before
being transmitted to the profession and the general public (Relman 1981). Ten
years after, the new editors Marcia Angell and Jerome P. Kassirer reinforced the
need for the rule to remain in place (Angell and Kassirer 1991).
Since its creation, the rule was adopted in most scientific journals. According to
Marshall (1998), an estimated 300 other journals follow guidelines similar to the
Ingelfinger rule, laid down by a group of medical editors calling themselves the
‘Vancouver group,’ a reference to their first meeting place in Canada, in 1978.
Also since its creation, the rule has gathered a chorus of voices for (Wilkes and
Kravitz 1995; Johnson 1998; Steinbrook 2000) and against it (Goodell 1977;
Altman 1996a; Shuchman and Wilkes 1997). In a survey of senior editors of all
269 leading medical journals published at least quarterly in the United States
(Wilkes and Kravitz 1995), 84 percent advocated discussion of scientific findings
with the press, but only in accord with the Ingelfinger rule (77 percent).
7
The scientific literature has established the term ‘Ingelfinger rule.’ However, it must be said that
S.A. Goudsmith, editor of Physical Review Letters, had established a similar rule nine years before
Ingelfinger: ‘In the future, we may reject papers whose main contents have been published
previously in the daily press.’ Goudsmit, S. A. (1960). "Publicity." Physical Review Letters 4: 1-2.