Comparative Failure in Science

Originally published in Science, 143(3610) (March 6, 1964): 1012-14

Barney G. Glaser, Ph.D., Hon. Ph.D.

Besides degree of clarity, another aspect of a demotion is
the relative failure which it indicates. The clearer the
forms of comparative failure, the more painful they are
likely to be. However, a certain degree of failure may
indicate nothing more than the lack of outstanding
success, while indicating moderate success.

A perennial problem for some scientists is their feeling of
comparative failure
as scientists. This problem becomes clearer if
we consider two major sources of this feeling that are inherent in
the vary nature of scientific work. (i) In science, strong emphasis
is placed on the achievement of recognition; (ii) the typical basic
scientist works in a community filled with “great men” who have
made important and decisive discoveries in their respective fields;
they are the acknowledge guiding lights. These esteemed
scientists, who have attained honors beyond the reach of most of
their colleagues, tend to become models for those who have been
trained by them or who have worked under them. As Eiduson has
put it in her recent psychological study of basic research
scientists “Scientists: are idols-oriented.”

To take these honored men as models is important for
training as well as for a life of research. During training, one
learns to think creatively. Emulation of these models results on
the internalization of values, beliefs, and norms of the highest
standard. This emulation of the great continues and guides the
scientist in his research work, however individual in style his
work may be.

But it is precisely here that a feeling of comparative failure
may arise. In emulating a great man the scientist tends to
compare himself with the model. He estimates how closely he has
equaled his model in ability to adhere to high standards of
research, to think of relevant problems, to create “elegant”
research designs, to devise new methods, to write clearly, to
analyze data. In addition, because of the strong emphasis on
attaining recognition for research contributions the scientist
perhaps will compare his own degree of success with his model’s
to gauge how he himself is doing. In using the great man’s
achievements and the recognition accorded him as criteria, the
scientist may be motivated to strive continually and
unremittingly towards greater heights. On the other hand, he
may see himself, over time, as a comparative failure for not
having attained a comparable amount of recognition.

Eiduson brings out the dynamics of this problem for

The model, then, is the ego ideal figure who represents
the ultimate position, and in fact, defines what a scientist
should do, how he should think, how he should act. By
comparison, everything else is inevitably of lesser worth

[italics mine]. We have seen the way scientists in this
group rebuke themselves as they become old, distracted,
sit on committees or government advisory boards, or
become administrators- and thus move away from the
ideal. From this picture it is obvious that the scientist is
hard on himself. He has a built-in, clearly marked scalar
system, along which attitudes and kinds of performances
are measured. When he moves away and deviates from
the pattern, he becomes a maverick, or a person who has
tossed aside the flaming torch.

Average Success

With this basic problem in mind, I recently made a study of
the organizational careers of basic research scientists, one
purpose of which was to ascertain the consequences, for the
scientist’s career, of receiving or not receiving an average amount
of recognition: At the time of this study, these scientists were
employed in a government medical research organization devoted
to basic research. This was a high-prestige organization from the
standpoint of scientists and was run much as though it were a
series of university departments. The study is relevant to this
discussion in showing something of the career history of basic
research scientists, who are today affiliated with high-prestige
organizations devoted to basic research. In these contexts
organizational scientific careers are still primarily dependant on
professional (not organizational) recognition.

By “average amount of professional recognition” I mean
supervisor’s favorable evaluation of the quality of the scientist’s
current research, and proper credit, through publication and
through acknowledgement in the publications of others, for his
contribution to the cumulative knowledge in his field. This
definition gives three major sources of recognition within reach of
the typical scientist; references from superordinate colleagues,
publication, and publication acknowledgements in the work of
others. This “average” degree of professional recognition is
attained by most of the country’s scientists at any one time and
by practically all scientists at one time or another. This degree of
recognition is in marked contrast to the highly regarded, and
restricted, high-prestige honors (in the form of awards, prizes,
grants, lectureships, professorships, and so on) that are part of
the professional recognition accorded those scientists who make
great and decisive discoveries- the “great men.”

Three general aspects of scientists’ careers were studied:
performance; security in, and advancement of, position;
compatibility with others, and satisfaction with one’s location in
science. With respect to performance, an average degree of
recognition was found basic to high performance. That is,
recognition maintained high motivation to advance knowledge,
and high motivation resulted in the scientist’s devoting more of
his own time to research; this, in turn, resulted in high-quality
scientific performance, as judged by the researcher’s closest
professional colleagues.

Since, of course, such performance on the part of many
individuals is the basis of organizational prestige; it was not
surprising to find the organization providing, in return, a stable
scientific career for a scientist who received average professional
recognition. The scientists accorded this degree of recognition, in
contrast to those accorded less, felt more satisfaction in their jobs
and salaries. They tended to be more optimistic about their
chances of promotion, and their rate of promotion was higher.
With respect to the conditions for research – a most important
consideration for basic-research scientists – they fared
considerably better than scientists not accorded average
recognition. They had more freedom to work on their own ideas,
had more chance for originality, had more chance to use their
current abilities and knowledge as well as to gain new abilities
and knowledge, and had generally better research facilities and
supplies. In sum, the “average” recognition accorded them was
sufficient to give them security and advancement in their
scientific careers.

Lastly, with average recognition, the high quality
performance and steady advancement could be achieved in a
setting that provided personal satisfactions. The scientists
accorded average recognition, again in comparison to those
accorded less, were more content with their research and nonresearch
colleagues. More of them felt intense interest in working
with close professional associates. They were more satisfied with
their assistants and with the other scientists, the organization
leaders, their own supervisors and the directors of their
particular institutes. They felt strengthened through belonging to
work groups, such as sections and laboratories. They depended
more on personal contacts for scientific information, both inside
and outside the organization. They participated more in
seminars, meetings, and the activities of professional clubs and
other small groups.

Closely linked with this compatibility with their associates
was a satisfaction with their location in the community of
organizations of science. The scientists accorded average
recognition, in comparison to those accorded less, felt strongly
attached to their respective institutes and organizations. Indeed,
they felt more satisfied with the organization’s reputation in the
scientific world, and more of them felt that a sense of belonging to
an organization which had prestige in both the scientific and the
general community was of utmost importance. In comparing their
own organization (from the standpoint of what job factors they
deemed most important) with the “best” universities, hospitals,
industrial research organizations, and government research
organizations, more of them consistently reported that their
organization was generally better. In sum, the context of their
careers in science was highly favorable.

Together these findings suggest that an average amount of
recognition has a generally stabilizing effect for the careers of the
scientists within the high-prestige organization of the study.
(Even for individuals who received little or no recognition, the
pressure on careers was not so great as to cause an exodus from
the organization or from science itself. The great majority of these
men thought the lack of recognition was only temporary and
planned to continue in the organization, trying to advance

These findings suggest that career stability based on average
professional recognition is probably found in other organizations
similar in nature to the basic-research organization of this study,
and that in organizations of lesser standing even less recognition
may assure career stability. In the light of these findings it
appears that the feeling of comparative failure that may result
when the average scientist judges his lesser success by the
considerable success of his “great man” model tends to occur in
many instances within the context of a stable, promising career.
Further, most scientists can gain, if they do not have it currently,
the degree of recognition necessary for a stable career.
Comparative failure, then, is an evaluation resulting from a
social comparison. It is not to be taken as an absolute failure (loss
of position as a scientist). A comparative failure can still be
successful; an absolute failure is through.

The Scientific Career: A carnivorous god?

Comparisons with great men are, however, taken not as
comparative but as absolute failure by Kubie in his famous “Some
Unsolved Problems of the Scientific Career.” Kubie warns future
scientists of the perils ahead when devoting themselves to that
“carnivorous god, the scientific career.” His criteria in warning of
potential failure are absolute (not comparative) judgments, based
on the careers of the more notable great men of science. For
example, he talks of the “ultimate gamble which the scientist
takes when he stakes his all on professional achievement and
recognition [italics mine], sacrificing to his scientific career
recreation, family, and sometimes even instinctual needs, as well
as the practical security and money.” Implying again that the
scientist whose success falls short of the great man’s is an
absolute failure, he characterizes the young scientist as having “a
self deceiving fantasy: that a life of science well may be tough for
everyone else, but that it will not be for him,” and as having
“ambitious dreams; unspoken hopes of making great scientific
discoveries; dreams of solving the great riddles of the universe.”

Kubie states that the young scientist “dreams unattainable
dreams.” More directly relating his judgments to great men, he
cautions against choosing science as a career, because of the
“many failures it took to make one Pasteur.” He states that most
young scientists, in using great men as models, unwittingly set
themselves up to become failures: “…most young men view the
prospect solely by identifying with the most successful chiefs,
never stopping to consider how many must fail for each one who
reaches this goal.” Without making the distinction between
absolute and comparative failure, this last statement clearly
implies the former.

Admittedly, from this standpoint man must fail and few will
attain the stature of their models, but this is hardly a reason for
dissuading young men from becoming scientists. The chance is
slight that they will equal or surpass their models, but they
should be informed that most can gain the fundamental degree of
recognition indicated in my study as necessary for a promising
career in science. Surely the career to which they commit
themselves need not be as Kubie says, “devoid of security of any
kind, whether financial or scientific.”

Furthermore, these young men should be encouraged to
enter science and take great men as their models, for most will be
the artisans who do the commendable, but not the earthshattering,
research which accumulates to form the foundation
for future decisive advances. Kubie himself has recently, although
somewhat ambivalently, recognized this, in comparing the typical
scientist with the internationally famous scientist. “These little
known and unrewarded men are the expendables of science. They
are no less essential than are the few who reach their goals.
Therefore, until many years had passed it would be hard to weigh
which of these two men had had the more profound impact on
scientific knowledge.”

Perhaps my discussion draws the kind of “implication” from
“statistics” that Kubie is looking for in future research when he
says in his article on the scientific career: “It is the…duty of
scientists and educators to gather such vital statistics on the life
struggles of a few generations of scientists and would-be
scientists and to make sure that every graduate student of the
sciences will be exposed repeatedly to the implications such data
may have for his own future.” Career decisions are perhaps
among the most important determinants of a man’s fate, and
anything which contributes to an understanding of the career in
science may help people make these decisions more wisely.

Research on Comparative Failure in Science

While it is possible to be a comparative failure in virtually
any occupation, the chances of becoming one are built directly
into any occupation of scientist. Yet little or nothing is known
about this area of comparative failure. Therefore, I wish to
discuss a few of the properties of comparative failure that may be
useful for guiding exploratory research. To be sure, the most
important and meaningful properties of this problem area are yet
to be discovered. Since comparative failure is based on some
social comparison, statements about it must always take into
account a reference criterion. In this study, the reference
individual is the “great man” model in science, and comparisons
are based on his degree of success. The criteria for this invidious
comparison must be established empirically for any occupation
within its particular situational context. The problem of
comparative failure may be seen as a specific one within the more
general area of comparative reference group theory.

The relative nature of comparative failure is in marked
contrast to the absolute nature of failure wherein one cannot hold
an occupational position and therefore must leave it.
Comparative and absolute failures vary independently. A person
who is an absolute may or may not be a comparative failure. It is
true that a scientist forced out of science because of mediocre
work will probably feel himself a comparative failure with
reference to his many former colleagues and models. However, if
he can leave these comparative reference individuals behind
during the “cooling-out” process attached to his loss of
occupational position, he may not at all feel himself a
comparative failure. More simply, if he realizes from his absolute
failure that he was not cut out to be a scientist, then evaluations
of comparative failure become superfluous and may not persist.
Again, comparative failure can unnecessarily cause absolute
failure; that is, cause a scientist to leave his profession or to
stumble along feeling that he is an absolute failure.

Comparative failure takes on more strategic meaning when
no absolute failure is involved. A major source of comparative
failure is, of course, demotion or downgrading of one’s rank in the
occupation. This topic has been analyzed quite nicely for business
and industrial organizations. It had yet to be considered for
scientists. Another general source of comparative failure appears
when an important reference individual outstrips the scientist.
Statements about colleagues, such as “He advanced very quickly,”
can mean that the speaker was probably left behind with a lesser
degree of success. Practically all scientists have classmates and
colleagues who have been far more successful.

Another interesting and significant source of comparative
failure was brought out in this volume. With reference to the
“great man” model, the typical scientist can have an objectively
satisfactory career and yet feel a comparative failure! If this
feeling persists, he may be oblivious to the stability of his
promising career. If he actually lacks the fundamental range of
recognition, and further, feels that his chances of gaining some
recognition are hopeless when compared to his model, in focusing
on the unattainable he is likely to ignore the general possibility of
having a fruitful (if lesser) career by achieving the degree of
recognition within reach of his ability.

The adjustment to comparative failure involves a “cooling
down” of aspirations for success so as to be in accord with one’s
research ability and career prospects (rather than a “cooling out,”
as Goffman has described it for absolute failure, when one must
give up all those aspirations linked with an occupation or
position). The scientist as comparative failure does not have to go
through the often very painful process of giving up commitment,
involvement, and investment in his chosen profession. He need
only set his achievement and career sights at a lower level than
those of his models. Indeed, his models, who provide the basis for
an unfavorable comparison, also provide a perspective on just
where to see hopes for a career in science. This perspective is
denied the absolute failure, since he is going nowhere.

Various aspects of both the scientist and his “great man”
model will affect judgments of comparative failure. Aspirations
for a success similar to the model’s probably are highest
immediately after graduate school, but they diminish as the
scientist takes his own and his model’s measure during the years
of research maturity that follow. This change most likely will
vary according to the age at which the model did outstanding
research; in some fields, great men appear at an earlier age than
in others. If he is in a field in which great discoveries come early,
the scientist may feel that he is “through” soon after he has
begun. If the discoveries come late, then he may believe he has
ample time to equal his model – and his feelings of comparative
failure may emerge much later.

The type of models that a scientist takes will also vary
according to his age. The young scientist usually is focused on his
classmates, equal colleagues, teachers, his work from supervisors,
and current great men in his field. Hence, he has many reference
individuals on which to base a feeling of comparative failure
before he cools down his aspirations. In later years, as these
models no longer loom so large, the more successful scientist may
switch to the non-living immortals of his field. Indeed, although a
scientist may have become a great figure, he may still evaluate
himself strongly as a comparative failure, judging himself in
relation to the immortals he aspires to equal. Notwithstanding
the many useful consequences of taking immortals as models, it is
quite possible that comparative failure can be even more intense
for a contemporary great man whose aspirations (in contrast to
those of the typical scientist) may have been sharpened, not
dulled, on the grindstone of experience. In other words, not only
may a comparative failure be successful, but the most successful
scientist may be the most intense comparative failure. Thus,
comparative failure may be more pronounced among the
beginning scientists and current great men than it is among

It should be noted also that I have dealt only with
comparative failure as self-evaluation, not as other-evaluation.
The latter also represents an area for research in the sociology of
science. For example, a strategic aspect of advancement in the
scientific career is the comparative evaluations given by the
scientist’s referees. These are speculations. Only future research
can tell us of the processes leading to comparative failure; of its
incidence at various stages of an objectively successful career; of
the “cooling down” mechanisms by which scientists cope with
such a feeling and are helped by others to cope with it: of its effect
on creativity, motivation, and partial retreatism; and of its effects
on absolute failure and organizational turnover.