1956 IRE TRANSACTIONS---INFORMATION THEORY 3
The Bandwagon
CLAUDE E. SHANNON
NFORMATION theory has, in the last few years, subject are aimed in a very specific direction, a
become something of a scientific bandwagon. direction that is not necessarily relevant to such
Starting as a technical tool for the communica- fields as psychology, economics, and other social
tion engineer, it has received an extraordinary sciences. Indeed, the hard core of information theory
amount of publicity in the popular as well as the is, essentially, a branch of mathematics, a strictly
scientific press. In part, this has been due to connec- deductive system. A thorough understanding of the
tions with such fashionable fields as computing ma- mathematical foundation and its communication
chines, cybernetics, and automation; and in part, to application is surely a prerequisite to other applica-
the novelty of its subject matter. As a consequence, tions. I personally believe that many of the concepts
it has perhaps been ballooned to an importance of information theory will prove useful in these other
beyond its actual accomplishments. Our fellow scien- fields-and, indeed, some results are already quite
tists in many different fields, attracted by the fanfare promising-but the establishing of such applications
and by the new avenues opened to scientific analysis, is not a trivial matter of translating words to a new
are using these ideas in their own problems. Applica- domain, but rather the slow tedious process of
tions are being made to biology, psychology, lin- hypothesis and experimental verification. If, for
guistics, fundamental physics, economics, the theory example, the human being acts in some situations like
of organization, and many others. In short, informa- an ideal decoder, this is an experimental and not a
tion theory is currently partaking of a somewhat mathematical fact, and as such must be tested under
heady draught of general popularity. a wide variety of experimental situations.
Although this wave of popularity is certainly Secondly, we must keep our own house in first class
pleasant and exciting for those of us working in the order. The subject of information theory has cer-
field, it carries at the same time an element of danger.
tainly been sold, if not oversold. We should now turn
While we feel that information theory is indeed a
our attention to the business of research and devel-
valuable tool in providing fundamental insights into
opment at the highest scientific plane we can main-
the nature of communication problems and will
tain. Research rather than exposition is the keynote,
continue to grow in importance, it is certainly no
and our critical thresholds should be raised. Authors
panacea for the communication engineer or, a fortiori,
for anyone else. Seldom do more than a few of should submit only their best efforts, and these only
nature’s secrets give way at one time. It will be all after careful criticism by themselves and their col-
too easy for our somewhat artificial prosperity to leagues. A few first rate research papers are preferable
collapse overnight when it is realized that the use of a to a large number that are poorly conceived or half-
few exciting words like information, entropy, redun- finished. The latter are no credit to their writers and
dancy, do not solve all our problems. a waste of time to their readers. Only by maintaining
What can be done to inject a note of moderation in a thoroughly scientific attitude can we achieve real
this situation? In the first place, workers in other progress in communication theory and consolidate
fields should realize that the basic results of the our present position.
