Turing’s Test 107
the White House”.
28
The unexpected behaviour on the President’s part
means that any attempt to provide a form of regular news algorithm is
doomed to failure. This is not to say that there are not generic codes and
patterns that have emerged in terms of reporting the White House, for
example the almost religious close reading that has taken place whenever
Trump tweets. Nonetheless, without even thinking of artificial automation of journalism, if the task of human-driven news algorithms is to – in
the famous words of Gaye Tuchman – “routinize the unexpected”,
29
then the activities of the 45th President have made that a much more
difficult procedure. Writing in 1973, Tuchman’s insight was that journalists were expected to give accounts of a wide range of unexpected
activities on a routine basis, and that the two important elements of this
activity revolved around how they identified such events as newsworthy
and how the rituals and routines of their workplace enabled them to produce their reports.
30
Drawing upon well-recognised categories such as
“hard” and “soft”, as well as “spot” and “developing”, news, Tuchman
demonstrated a basic formal structure by means of which organisations
could schedule non-timely (usually “soft” news) and prescheduled events
such as debates and press conferences. This semi-structural approach
to the production of news had been taken a stage further in Galtung’s
and Ruge’s 1965 paper on the structure of foreign news,
31
which produced a famous list for classifying the newsworthiness of events that is
much debated and even frequently used today, for example Harcup and
O’Neill
32
and Joye, Henrich, and Woehlert.
33
Such structuralist and
organisational approaches to the production and dissemination of the
news by people remain important, as when Moon and Hadley discuss
the routinisation of Twitter as a source for gathering stories
34
or Hartley demonstrates how Danish breaking news is regularly categorised.
35
While the responses of Trump may be less easily predictable than former presidents, even to the extent that White House press briefings have
taken a very different form since his inauguration in 2017, nonetheless
there remains a press corps attached to the White House, with prescheduled events occurring even if it is not always clear what will be said
or done in advance. The unpredictability of Trump has been extremely
good in terms of injecting brio into North American journalism (and,
indeed, for that in many other places around the world), but that unpredictability is also made possible because so much of what happens still
takes place in the orbit of Washington.
Overall, it seems unlikely that political reporting in the United States
is likely to be automated any time soon. Indeed, as we shall see in the
next chapter, the fact that editorialising in relation to the President
rather than simply reporting his actions has become the new norm
makes such automation extremely difficult if not impossible. The kinds
of highly structured news stories considered previously in this chapter simply lack the format of data required to convert them into copy,
leaving aside Trump’s propensity for distorting facts to suit his current
the White House”.
28
The unexpected behaviour on the President’s part
means that any attempt to provide a form of regular news algorithm is
doomed to failure. This is not to say that there are not generic codes and
patterns that have emerged in terms of reporting the White House, for
example the almost religious close reading that has taken place whenever
Trump tweets. Nonetheless, without even thinking of artificial automation of journalism, if the task of human-driven news algorithms is to – in
the famous words of Gaye Tuchman – “routinize the unexpected”,
29
then the activities of the 45th President have made that a much more
difficult procedure. Writing in 1973, Tuchman’s insight was that journalists were expected to give accounts of a wide range of unexpected
activities on a routine basis, and that the two important elements of this
activity revolved around how they identified such events as newsworthy
and how the rituals and routines of their workplace enabled them to produce their reports.
30
Drawing upon well-recognised categories such as
“hard” and “soft”, as well as “spot” and “developing”, news, Tuchman
demonstrated a basic formal structure by means of which organisations
could schedule non-timely (usually “soft” news) and prescheduled events
such as debates and press conferences. This semi-structural approach
to the production of news had been taken a stage further in Galtung’s
and Ruge’s 1965 paper on the structure of foreign news,
31
which produced a famous list for classifying the newsworthiness of events that is
much debated and even frequently used today, for example Harcup and
O’Neill
32
and Joye, Henrich, and Woehlert.
33
Such structuralist and
organisational approaches to the production and dissemination of the
news by people remain important, as when Moon and Hadley discuss
the routinisation of Twitter as a source for gathering stories
34
or Hartley demonstrates how Danish breaking news is regularly categorised.
35
While the responses of Trump may be less easily predictable than former presidents, even to the extent that White House press briefings have
taken a very different form since his inauguration in 2017, nonetheless
there remains a press corps attached to the White House, with prescheduled events occurring even if it is not always clear what will be said
or done in advance. The unpredictability of Trump has been extremely
good in terms of injecting brio into North American journalism (and,
indeed, for that in many other places around the world), but that unpredictability is also made possible because so much of what happens still
takes place in the orbit of Washington.
Overall, it seems unlikely that political reporting in the United States
is likely to be automated any time soon. Indeed, as we shall see in the
next chapter, the fact that editorialising in relation to the President
rather than simply reporting his actions has become the new norm
makes such automation extremely difficult if not impossible. The kinds
of highly structured news stories considered previously in this chapter simply lack the format of data required to convert them into copy,
leaving aside Trump’s propensity for distorting facts to suit his current