Objectivity,honesty,and integrity: How American scientists talked about their virtues,1945–2000

Abstract

What kind of people make good scientists? What personal qualities do scholars say their peers should exhibit? And how do they express these expectations? This article explores these issues by mapping the kinds of virtues discussed by American scientists between 1945 and 2000. Our wide-ranging comparative analysis maps scientific virtue talk across three distinct disciplines – physics, psychology, and history – and across sources that typify those disciplines’ scientific ethos – introductory textbooks, book reviews, and codes of ethics. We find that, when inducting students into a discipline, evaluating peers, or codifying their professional standards, postwar American scientists routinely named virtues like carefulness, objectivity, and honesty. They applied such virtues not only directly to scholars’ characters, minds, and attitudes (thereby equating virtues with personal qualities), but also to their methods, modes of reasoning, and working habits (in the form of what we call virtue-qualifiers). Strikingly, we find that physicists, psychologists, and historians drew upon largely similar repertoires of virtue. For all of them, scientific work required carefulness, thoroughness, and accuracy. Not all virtues, however, were equally important in all disciplines (notably objectivity), nor did each ethos-forming genre place equal emphasis on the directly personal nature of such virtues. All in all, our research establishes an extended framework for understanding the ways virtues remained present in postwar American scientific discourse writ large.

Introduction

What kind of person makes a good scientist? Does scholarship demand certain personal qualities of its practitioners? Prior to the twentieth century, these were routine questions, recognizable to every scholar. As Steven Shapin and others have demonstrated, early modern natural philosophers frequently talked about virtues relevant to the life of the mind. ¹ Not only were they convinced that scholarship required character traits like impartiality, humility, and perseverance, but they also believed that scholarly activity had an ennobling effect on the character of its practitioners. Learning, in other words, had a doubled relationship to virtue: it both demanded and fostered virtuous ways of being and acting. ² Newer research on scholarly virtues and vices has shown that the first belief, that good scholarship requires virtue, remained of vital importance throughout the eighteenth, nineteenth, and early twentieth centuries. ³ While the second belief – science making better people – gradually disappeared into the background, the importance of certain qualities for doing science continued to receive vigorous attention, especially as fields of learning established themselves as academic disciplines. ⁴ Professionalization prompted reflection on the differences between scientists and amateurs as well as on the possibilities and limitations of scientific education (to what extent could scholarly virtues be taught?). ⁵ Meanwhile, scholarly virtues (“objectivity”) and vices (“dogmatism”) played key roles in post-Darwinian controversies and other debates at the intersection of science, politics, and religion. ⁶ For the United States, it has been shown that virtues were a privileged idiom for evaluating scholarly accomplishments until at least the 1930s or 1940s. ⁷ We can ask, however, to what extent this changed later in the century.

Where existing scholarship touches on this question, it does so by identifying two interrelated developments. ⁸ First, historians such as Andrew Jewett have observed that scholars in the 1920s and 1930s United States replaced the notion of a “scientific spirit,” defined in terms of personal virtues, with that of a “scientific attitude.” ⁹ Arguably, this does not tell us much about the vicissitudes of virtue terms like impartiality, honesty, and accuracy. In earlier centuries, virtue terms had also been grouped under different labels, such as “duties,” “tasks,” and “responsibilities.” Secondly, it has been argued that the declining popularity of “virtue” as an ethical umbrella term, combined with “the disappearance of the individual” and the instrumentalization of science for political and military purposes, caused the ties between science and virtue to wither. ¹⁰ On the one hand, this would have meant that “by the middle of the twentieth century, explicit talk about virtue had largely disappeared from discussions of science,” as Thomas Stapleford and Emanuele Ratti propose. ¹¹ Where this refers to scientists’ use of virtue terms like honesty and accuracy, however, the argument still needs to be tested empirically: scholars have not yet provided substantial evidence for or against (a gap that this article aims to fill). On the other hand, the literature is replete with master narratives of modernization, bureaucratization, and secularization that tend to associate virtue with the prehistory of modern science – a time when learning was still largely an individual, religiously inspired pursuit – more than with twentieth-century science. What such narratives fail to appreciate, in the view of Steven Shapin, is that “people and their virtues” continued to matter “to the making and the authority of late modern bodies of technical knowledge,” even or precisely in the most advanced branches of American technoscience. ¹² In sum, the theme of scholarly virtues in the twentieth century has been touched upon, but not nearly with the degree of empirical rigor that scholarship on earlier periods has attained. This paper begins to fill this gap by mapping routine discourse about virtues across three scholarly genres in disciplines spanning the full range from the humanities to the physical sciences.

For a historical study of scholarly virtues in postwar science – preferably in the United States, given the global prominence of American science in the postwar period – two things seem crucial: a clear definition of terms and a research design that encompasses more than isolated case studies. ¹³ Regarding definitions, we use the term “virtues” as synonymous with personal qualities, character traits, or habits of mind that scholars invoked when discussing the personal demands that science made on its practitioners. “Scholarly virtues,” then, are specifically qualities, traits, or habits that scholars qua scholars were expected to possess, cultivate, or display. ¹⁴ This is not a definition likely to be found among virtue ethicists or virtue epistemologists, who usually work with more narrowly construed accounts of moral and intellectual virtue. ¹⁵ Our definition does, however, align closely with historical scholarship on earlier periods, which has recognized the fuzzy boundaries between the epistemic and the ethical in scholars’ talk of virtue as well as scholars’ constant negotiation, in terms of virtues and vices, of political and religious demands made on research and teaching. ¹⁶ “Scholarly virtues” is consequently an analytical category rather than an actor’s term: it denotes a historically evolving set of personal qualities (accuracy, honesty, impartiality, loyalty, thoroughness, objectivity, open-mindedness, and so forth) that scholars invoked in specifying the characteristics of a good scholar. ¹⁷ What we will map, then, is not the changing popularity of the umbrella term “virtue,” but the configurations of specific virtue terms (objectivity, honesty, integrity, etc.) in postwar scientific discourse.

We have two reasons for focusing on scientific discourse about virtues – virtue talk, as we call it – rather than on personal qualities that scientists actually possessed or displayed. First, while there is no lack of sources referring, for instance, to Albert Einstein’s curiosity or Thomas Huxley’s dogmatism, such attributions of virtue and vice usually tell us as much about the writers as about the individuals discussed. Similarly, although biographies and obituaries are notoriously unreliable when it comes to the virtues they ascribe to past scientists, such genres do showcase how virtues and vices served as a vocabulary for articulating ideals and expectations regarding scholars’ conduct and working manners. ¹⁸ Secondly, as Lorraine Daston and Peter Galison remind us, such idioms have histories: they evolve over time. ¹⁹ New virtues (open-mindedness, fair-mindedness) might be added, while others (piety, humility) could lose the prominence they once had. ²⁰ There are multiple historical examples of scholars engaging critically with such shifting valorizations. ²¹ Indeed, controversies about the relative weight of individual virtues have historically been the rule rather than the exception, if only because virtues like patriotism and loyalty (to the nation, the church, or the cause of democracy) stood in tension with more ascetic virtues like impartiality and objectivity. ²² Virtue talk, therefore, must not be mistaken for a coherent system of moral standards: it served as an idiom for articulating a whole range of sometimes irreconcilable expectations.

The second crucial point that marks our study is its ambitious scope. Focusing on American science in the second half of the twentieth century, we investigate discourse about virtue across disciplines spanning the arc from the natural sciences to the humanities. Rather than studying an isolated context or virtue, we provide the first (to our knowledge) bird’s-eye view of routine discourse about virtue across widely distinct disciplines and across a fifty-five-year period. With this extensive historical map, we gain a sense of the whole constellation of virtues that regularly mattered in the postwar American sciences, social sciences, and humanities. Where previous scholarship on twentieth-century virtues has mostly focused on single virtues (open-mindedness, objectivity), single scientists (Merle Tuve, Edwin Boring), or single disciplines (American historiography), our investigation provides a previously lacking framework for connecting and situating such individual, in-depth studies. ²³ It should be noted that we will often use the term “science” or “scholarship” as shorthand for “sciences, social sciences, and humanities,” and similarly refer both to “scientists” and “scholars.” ²⁴

Since our map aims to inform and orient future studies of individual virtues in discourse and practice, our approach in this article is deliberately broad and multidisciplinary in scope, exploratory rather than exhaustive in nature. The article proposes to encompass the full repertoire of virtue talk – which virtues were most regularly invoked, and how this varied between disciplines and across time – by examining places where the ethos of science was configured and transmitted as part of routine scholarly activity. Specifically, we focus on three genres in which scholars reflected on their disciplinary ethos or scientific norms: introductory textbooks of the kind read by first-year students, book reviews in scholarly journals, and ethics codes issued by major professional/scholarly societies. ²⁵ We study these genres in three academic disciplines: a prominent postwar social science – psychology – and two fields more frequently the subject of comparative histories of (epistemic) virtue – physics and history. ²⁶ That psychologists often aspired to the status of the natural sciences, while American historians were variously housed in social science or humanities faculties, implies that our study also implicitly makes connections across typical scholarly groupings.

In examining the dynamics of virtue talk across these disciplines, we notice that in the period under review (1945–2000), virtue terms were no longer solely attributed to individual persons, as had been common in earlier centuries, but were also associated with ways of working, reasoning, or writing. To mark this difference, we speak, on one hand, about personal qualities if virtues – like carefulness – were attached or ascribed directly to individuals. On the other hand, if carefulness was attributed to an activity (measuring temperature) or to scientific output (a monograph), we speak about virtue-qualifiers. The claim that an experiment performed by Henry Rowland in 1879 was “a model of careful experimentation” differs crucially from the statement that Rowland himself was a model of carefulness. ²⁷ In the former statement, “careful” appears less as a distinct character trait than as a method or working habit, while in the latter, carefulness becomes a habit of mind solidly ingrained in the person of Rowland. The first “careful,” in our definition, is thus a virtue-qualifier, part of an “adjectival and adverbial arsenal” of virtue terms that are used, not to refer to an individual’s character, personality, or attitude, ²⁸ but to describe how scientists conduct their studies – how they perform an experiment, analyze a source, or reason out a conclusion. ²⁹ The difference is subtle but important: “carefulness” as a noun has a characterological force that the adjective “careful” lacks. Moreover, whereas the noun draws attention to itself, the adjective qualifies something else. To distinguish between personal qualities and virtue-qualifiers notably matters because it can help unpack the often sticky question of whether virtues are applied differently between the scholar and the scholarship. ³⁰

We attend to the prevalence and function of both kinds of virtue talk in three successive sections of this paper, with an eye to personal qualities that are presented as necessary or desirable for scientific performance and to virtue-qualifiers linked to scientific activities. Each section focuses on a particular genre – introductory textbooks, scholarly book reviews, codes of ethics – and compares virtue talk across the disciplines of physics, history, and psychology. While all three genres are sites for communicating and fashioning disciplinary and scientific identities, each has its particular audience and aims. For coherence, we focus on texts linked to American professional scholarly societies and their major review journals: the American Psychological Association (APA) and Contemporary Psychology (CP; later PsycCRITIQUES); the American Historical Association (AHA) and The American Historical Review (AHR); the American Physical Society (APS) and Physics Today (PT). For each discipline, we also purposively sampled a selection of U.S. introductory textbooks (see Table A1). Where possible, we have selected our corpus to span the decades from the end of the Second World War to the end of the century (i.e. about 1945–2000). ³¹ Thus, we have studied all codes of ethics issued by the three professional societies (see Table A2), although only the APA promulgated ethical standards from the 1950s. Next, we read book reviews from the first issue per year for each of AHR and CP, and the first five issues per year of PT, for an arbitrary year per decade. ³² Our analysis of U.S. introductory textbooks, finally, draws on a corpus of twenty-one widely used textbooks in physics, psychology, and history, with consideration of multiple editions where available, for a total of thirty-eight texts. ³³

From a close reading of relevant passages, we collate the most frequently invoked virtues and compare them across fields and time. Given that our aim is to map the ways in which virtues were called out in twentieth-century American science, we are not concerned with how (or whether) any of these virtues were enacted in scientific practice. Our discursive map should, however, provide a solid framework from which to address such important questions in future studies. Nor do we make any claim to chart comprehensively or quantitatively how any named virtue occurred over time or genre. In our penultimate section, we nonetheless offer some brief reflections on the changing meanings and importance attached to “objectivity” and “integrity.” Overall, we find that psychologists, historians, and physicists regularly mobilized personal qualities as they inducted students into a discipline, evaluated their peers, and encoded a certain image of their discipline in ethical standards. Our sources, however, reveal a preference for using virtue-qualifiers, especially in book reviews and textbooks, which in turn suggests further important twists to be explored in a longue durée history of scholarly virtues. ³⁴ Our distinction between personal qualities and virtue-qualifiers notably seems promising in this regard.

Introductory textbooks

Student textbooks are a good place to start. As a genre aimed at initiating first-year students “into the well-established views and practices of specific scientific communities,” textbooks typically included some discussion of what research in the discipline looked like and how (aspiring) researchers were expected to behave. ³⁵ As Ernest Hilgard’s Introduction to Psychology (1953) put it straightforwardly: “The first question a prospective psychologist must ask himself is this: Am I suited by abilities and interests to become a psychologist?” ³⁶ In a different idiom, but with similar questions in mind, other authors dwelt on the “qualities,” “personal traits,” “scientific temper,” “scientific attitude,” “habits of mind,” or “skills” required from researchers. ³⁷

Evidently, therefore, reference to individual qualities remained relevant, as can be seen through a comparison of successive editions (1953, 1962, 1971, 1983, 1996) of Hilgard’s psychology textbook. ³⁸ Already in the first edition, readers were told that psychology’s scientific ambitions did not allow for “partisanship,” “loyalty to a position held by one or another prominent teacher,” or “doctrinaire” commitment to a favorite theory. ³⁹ Subsequent editions repeated these warnings, meanwhile adding “stubborn prejudice” and “dogmatism” to the list of vices. ⁴⁰ At the same time, they encouraged aspiring psychologists to develop virtuous habits. For instance, the 1962 edition depicted “the objective psychologist” as someone with an “open mind,” “widely modest about what has been achieved so far,” and at times even engaging in “creative thinking.” ⁴¹ Later editions emphasized the importance of “caution” and “careful observation,” while granting the classic virtue of “objectivity” a central place in the moral economy of science: “Objectivity lies at the center of the scientific attitude.” ⁴² As if this did not sufficiently illustrate what demands psychological research placed on scientists’ virtues, the 1996 edition made the point explicit by speaking in the first person plural about “our” unvirtuous habits: “We are biased to see other people’s actions as caused by their internal traits . . . This bias often leads us into error.” ⁴³ Clearly, personal qualities were a matter of concern in all editions (in 1996 perhaps even more than in 1983, in response to a growing interest in “scientific integrity” – a theme to which we return in later sections).

This interest in the scientific self and its defining traits was not unique to the field of psychology: it appeared with equal intensity in history textbooks, and somewhat less prominently in physics textbooks. Most notably, some virtues were advocated in all three disciplines. Almost every title in our sample presented “carefulness” as an indispensable attribute of responsible researchers. Francis Sears and Mark Zemansky’s University Physics (1955, 1976, 1987), for instance, abounds with phrases like “note carefully,” “careful attention must be paid,” “it is essential to distinguish carefully,” and “we must always be careful.” ⁴⁴ A similar insistence on carefulness permeates David Halliday and Robert Resnick’s Physics for Students of Science and Engineering (1960, later reissued as Fundamentals of Physics). Over and over, the book uses the adjective “carefully” to specify how rules must be applied, systems defined, curves studied, frequencies chosen, and temperatures measured. ⁴⁵ Historians also agreed on the importance of carefulness. Apart from emphasizing repeatedly the need for “careful, thoroughgoing study,” which in practice largely meant “carefully” sifting primary sources and “carefully” testing the reliability of each and every statement, history textbooks summoned their readers to be careful in drawing conclusions “so as not to outrun the evidence.” ⁴⁶ Interestingly, the “careful use and interpretation” of source material was not their only concern: they also recommended “great care and patience . . . in achieving proper spelling and punctuation.” ⁴⁷ Even more practically, students could learn how to read secondary literature – quickly and selectively if possible, but slowly and carefully when an author’s arguments were at stake – and how to take notes (“be careful to distinguish between your summary and a direct quotation”). ⁴⁸ Carefulness was, in other words, a shared virtue that textbook authors in all three fields wanted their readers to develop.

Carefulness did not, however, occupy an equally important place in what we might call the fields’ constellations of virtues. While nobody questioned that carefulness mattered, the relative weight that textbooks attached to it differed significantly both across and within disciplines. While many psychologists resembled physicists by stressing the importance of “careful research” and “careful observation,” ⁴⁹ they distinguished themselves from their colleagues in physics by insisting also on objectivity as a defining mark of “the critical, analytical attitude characteristic of the scientific method.” ⁵⁰ The 1958 edition of Ruch’s book, for example, stated that terms must be “defined objectively” and measurements made “as objectively as possible.” “Objective observation” apparently was a sine qua non of psychological research. ⁵¹ Later editions elaborated on the theme, noting the “objective, unbiased attitude” or the “objectivity and impartiality” that psychologists must possess. ⁵² Concomitantly, Ruch warned students against vices that could detract from objectivity: “prejudices,” “interpretative bias,” and “premature, incomplete, inaccurate conclusion-drawing.” ⁵³ Clearly, this emphasis on objectivity, sometimes paired with “scientific rigor,” was an attempt to bolster the scientific status of the field. ⁵⁴

Just like psychology textbooks, manuals on historical method valorized both carefulness and objectivity. Their ideal researcher similarly exhibited accuracy and thoroughness, but also (at least to some degree) the virtue of objectivity or impartiality. The importance of accuracy and attention to detail is especially prominent in Gilbert Garraghan’s Guide to Historical Method (1946). ⁵⁵ Similarly, the 1969 edition of Louis Gottschalk’s Understanding History portrayed the “painstaking historian” as someone “with an eye for accuracy,” eager to “set down carefully the story he has extracted from the sources.” ⁵⁶ These virtues reflect a philological ethos of the kind that source-oriented historians had been cultivating since the mid-nineteenth century. ⁵⁷ At the same time, history textbooks emphasized a need for objectivity. Responding either directly or indirectly to Charles Beard’s dismissal of objectivity as a “noble dream,” most of them added that this quality was as valuable as it was difficult to attain. ⁵⁸ Garraghan, for instance, wanted historians to develop “a detached and neutral attitude,” in full awareness that absence of all prejudice was “a psychological impossibility.” ⁵⁹ Gottschalk added that objectivity was an “obligation,” even if historians “may rarely succeed” in reaching it. ⁶⁰ Another textbook spoke in gradual terms about “maximizing objectivity” and “minimizing subjectivity”: “We cannot eradicate all bias, but we can do much to minimize it.” ⁶¹ Although the consensus about this compromise changed over time, what matters for now is that historians, like psychologists but unlike physicists, considered objectivity just as important a virtue as carefulness or accuracy. Constellations of virtues could take on different forms in different disciplines (as well as in different subregions of a field).

Finally, we return to the question of whether all this virtue talk was about personal qualities or virtue-qualifiers. Were objectivity, honesty, and thoroughness portrayed as traits of personality that researchers had to cultivate? Insofar as these terms were used in noun form, the answer is largely affirmative. “The spirit of objectivity” was unmistakably a personal quality, as it referred to what Robert Daniels called the capacity of “recognizing and regulating one’s own opinions and biases.” ⁶² Similarly, adjectives such as “biased” and “unbiased” could be ascribed to the scientist as individual (as in “unbiased observer” and “an individual scientist may very well be biased”), just as “being sensitive” appeared as a personal character trait. ⁶³ When Ruch argued that even the most careful research design cannot prevent “a poor interviewer” from biasing the results, he made clear that researchers’ personal qualities were a matter of irreducible importance. ⁶⁴ Other adjectival forms, in contrast, were less frequently a matter of a scientist’s personal habits. Occasional mention of “the objective psychologist” or “an objective, unbiased attitude,” ⁶⁵ for instance, is outweighed in textbooks by references to “objective observation,” “objective data,” “objective measurement,” “objective methods,” “objective validation criteria,” and “objective knowledge.” ⁶⁶ Likewise, the adjective “accurate” was predominantly associated not with researchers but with measurements, experiments, predictions, and laboratory notebooks. ⁶⁷ In most cases, therefore, virtue talk in textbooks occurred under the sign of virtue-qualifiers: it described other aspects of science than the person of the researcher, such as methods (as in “precise measurements,” “accurate experiments,” “careful manipulation,” and “carefully controlled research”), ⁶⁸ modes of reasoning (as in “prove rigorously” and “defining things precisely”), ⁶⁹ and working habits (as in “carefully observing,” “rigorous understanding,” and “study these curves carefully”). ⁷⁰

Book reviews

Virtue-qualifiers similarly abound in book reviews, a genre aimed at informing scholars about “the current scene [in a discipline] by telling its readers what the recent books contain” and evaluating the significance of those books. ⁷¹ Notably, the three journals studied here had the additional aim of fostering “a sense of professional unity and morale” in their respective disciplines by providing a space for researchers from different (sub)specializations to learn about “general trends in fields outside their own.” ⁷² Book reviews played a central role in furthering this goal. Contemporary Psychology was entirely comprised of reviews, while a typical issue of The American Historical Review would feature several hundred book reviews to some four or so research articles. ⁷³

In his instructions to book reviewers, Edwin G. Boring, the first editor of CP, pointed to the importance of evaluating books instead of their authors’ habits. “Criticize the text, the ideas, the logic, the accuracy, not the author,” he wrote. “Let all criticism be ad verbum, never ad hominem.” ⁷⁴ For the most part, reviewers complied with this request, not only in CP, but also in AHR and PT. In pointing to “fair-minded . . . work” or “careful and meticulous explanations,” they applied virtue adjectives either to a book’s contents – its analysis, explanation, or interpretation – or to its author’s mode of reasoning, working habits, or methods (translation, transcription, experimentation, interviewing). That is, virtues appeared not in the form of personal qualities but in the form of virtue-qualifiers. ⁷⁵

Most common in our sample of book reviews were derivatives of the virtues of “carefulness” and “thoroughness,” notably “careful” and “thorough.” In CP and PT, books were lauded for their “careful and thorough . . . assessments” of patients or for being “a careful, thorough statement.” ⁷⁶ Conversely, they were criticized for containing “uncritical discussions” of quantum theory presented in an “exaggerated style of writing,” with reviews across the three disciplines enlisting variations of virtue-qualifiers of this kind: “balanced,” “nuanced,” “original,” “critical,” and “fair.” ⁷⁷ Historian Stanley G. Payne, reviewing a book on the history of Spanish Catholicism, invoked several of these in just a few lines. He praised the author, Frances Lannon, for a work that was not only “valuable and original” but also “sensitive and carefully nuanced.” Moreover, Payne noted that while “Lannon’s analysis is often acerbic and highly critical, [it is] not lacking in basic fairness and judiciousness.” ⁷⁸

Some of these virtue-qualifiers were discipline-specific. For example, while “judicious” was a recurring virtue-qualifier in AHR, it was largely absent from the other journals. ⁷⁹ Disparities between disciplines were especially pronounced, however, when it came to authors’ personal qualities. Notwithstanding Boring’s warning that “CP does not provide space for the discussion of the intelligence or integrity of an author . . . personal aspersions are taboo,” postwar American psychologists sometimes critiqued authors as individuals explicitly and directly. ⁸⁰ In history as well as in psychology, but not in physics, there was a systematic emphasis on the “empathy” of authors toward their subjects, colleagues, and readers, as well as on the related virtues of “compassion” and “sympathy.” A biographer of Léon Blum was praised not only for being “a meticulous researcher and a fine craftsman,” but especially for “combin[ing] rigorous fairness with a deep-felt sympathy for his protagonist.” ⁸¹ Decades later, Bernardo Carducci was praised in CP by the Stanford prison-experimenter Phillip Zimbardo for approaching the topic of shyness “authoritatively” yet “compassionately,” thereby appealing effectively to his intended audience of “shy people.” ⁸² Yet if it was chiefly considered positive to affirm that a historian had displayed “warm sympathies for the men about whom he writes,” ⁸³ or psychologists had “written with an overwhelmingly therapeutic, pastoral, and compassionate orientation,” there were also book reviewers who insisted that an excess of such personal qualities had a distorting effect on historical and psychological scholarship. ⁸⁴ An author could have an “oversympathy for his subject,” as one AHR reviewer complained. ⁸⁵ Similarly, in the first issue of CP, an author’s excessive “adoration [and] compassion” for Sigmund Freud was presented as a vice that carried him “headlong through the most obstinate counter evidence when he [was] in the mood to claim another scrap of territory for the Master.” ⁸⁶

Besides “empathy” and its cognates, “objectivity” was a virtue invoked by book reviewers in history and psychology alike. Objectivity was said to be a “laudable” personal quality, associated with “candor,” “wisdom,” and “courage.” ⁸⁷ The simultaneous appearance of empathy and objectivity as virtues in book reviews in AHR and CP suggests that scholars held somewhat inconsistent, or even conflicting, views on the attributes of the ideal historian and the ideal psychologist. According to many historians and psychologists, however, the virtues of objectivity on one hand, and compassion, sympathy, or empathy on the other, were actually compatible when neither was taken to excess. Thus, psychologists judged their colleagues on whether they had managed to balance “their objectivity with the need to be appropriately compassionate,” or to “walk the fine line between being detached scientists and being sympathetic observers.” ⁸⁸ Historians sought to achieve a similar balance between an empathic and a detached scholarly attitude, as was the finding of a review in AHR of Bonnie Smith’s The Gender of History. In the review, Smith’s scholarly attitude toward the work of nineteenth-century women historians was praised for being “respectful . . . but, ultimately, dispassionate.” ⁸⁹

Given the tight link that historians and psychologists established between objectivity (or “dispassion”) and “scientific” method, it is remarkable that the virtue of “objectivity” was virtually absent in book reviews in physics. This accords with what we find for introductory textbooks; there seems to have been little need to assert the “scientific” status of physics in the Cold War context. ⁹⁰ The reviews published in PT occasionally mention other personal qualities, however, including “imagination” and “enthusiasm,” as well as “dogmatism” and “criticism.” In 1967, for instance, T. Teichmann emphasized that readers of An Introduction to Probability Theory and its Applications “will be grateful to [author William] Feller for his perseverance and enthusiasm in completing his venture,” while British–American quantum theorist Freeman Dyson criticized the authors of a book on quantum field theory for lacking “a critical and enquiring mind.” ⁹¹

That the naming of personal qualities was very much present in postwar U.S. academic book reviews is further illustrated by the many instances in which book reviewers explicitly evaluated the “minds” of their fellow scholars. Minds could be praised for being “broad and penetrating” and “relentlessly probing [and] sensitive,” or, as in Dyson’s review, criticized for being insufficiently “critical.” ⁹² In addition, book reviewers explicitly criticized authors’ attitudes, and diagnosed some psychologists with a “partisan attitude,” or, worse, a “cavalier attitude towards evidence.” ⁹³ Historians’ attitudes, in turn, were praised for exhibiting maturity and responsibility. ⁹⁴

Codes of ethics

Scholars’ minds were also at issue in professional codes of ethics. Although ethics codes have been criticized as “symbolic resources” that express “power relations” in a profession, ⁹⁵ they are also valuable (and understudied) sources insofar as they offer glimpses of the attitudes or comportments that made up a field’s self-image. ⁹⁶ This self-image often came with specific expectations of practitioners, as is apparent from the prominence of personal qualities in the three societies’ ethics codes. Thus, it is the only one of our three genres to give an equally important, if not more important, place to personal qualities as to virtue-qualifiers. Almost all the codes in our corpus begin by naming some personal quality – the exception is the first set of ethical standards issued by the APA (in 1953), which notably framed its nearly 200 pages around ethical “problems” and example “incidents.” The APS, for instance, began its guidelines for physicists’ professional conduct with the personal quality of honesty: the first paragraph elaborates on the notion that “honesty must be regarded as the cornerstone of ethics in science.” ⁹⁷ This prominence of honesty in the physicists’ ethics code is all the more striking as the APS only promulgated one such document during the twentieth century (in 1991), of comparatively brief length. Other personal qualities also make an appearance in these few paragraphs, notably objectivity, effectiveness, and (professional) integrity, which are held to characterize the physicist engaged in ethical behavior. A situation involving a “conflict of interest” can be recognized, according to the APS, by the absence of these qualities. ⁹⁸

Of the four personal qualities that characterized the ethical physicist, it is on integrity that various APA and AHA ethics codes repeatedly insisted. Somewhat surprisingly, given our findings in regard to textbooks and book reviews, objectivity was not central to the way psychologists or historians framed their “core set of shared values.” ⁹⁹ Integrity, in contrast, headed principles in the 1974 AHA and 1992 APA ethics codes, merited mention in the preamble to the 1959 APA code, and was emphasized (in bold or italics) in both research and teaching sections of the 1987 and 1999 AHA professional standards. ¹⁰⁰ In other words, integrity was labeled prominently as crucial to the make-up of a good historian, and relatively consistently to that of a good psychologist.

But if cultivating integrity certainly made for good scholarship, can we categorize it as properly a virtue? Always a noun attached to the individual historian, psychologist, or physicist, does it count as a personal quality in the sense of a habitual trait of personality or character? Several APA ethics codes point toward answering in the affirmative. “He maintains integrity with respect to the facts of his science and in his relationships with other psychologists and with the public,” declared the preamble to the 1959 code, while the 1992 revision has psychologists as seeking to “promote integrity in the science, teaching, and practice of psychology,” under the principle titled “integrity.” ¹⁰¹ Here, integrity appears as a quality that mediates or configures a psychologist’s approach to various aspects of his (not yet her) scholarly activity. Like an Aristotelian virtue, it would seem to induce its possessor to act in the right way. If interpreted in these terms, “integrity” would join the constellation of scholarly virtues alongside the more traditional honesty and objectivity. ¹⁰²

A somewhat different interpretation emerges, however, if we read on further in the 1992 APA code or analyze statements by the AHA from 1974 onward about “integrity in research” and “integrity in teaching.” The principle of integrity means that “psychologists are honest, fair, and respectful of others,” according to the 1992 APA ethics code. ¹⁰³ For its part, the AHA glossed “integrity in teaching” along similar lines as a matter of “intellectual honesty,” “fairness in judging students’ work,” and “open-mindedness” in discussions with students. ¹⁰⁴ Among other things, “integrity in research” asked of historians “an awareness of one’s own bias and a readiness to follow sound method.” ¹⁰⁵ Such statements construe integrity less as an autonomous personal quality and more as an umbrella virtue that groups together a range of personal qualities (honesty, awareness) and even some virtue-qualifiers (fair judgment).

“Competence” takes on much the same dual characteristics in historians’ and psychologists’ ethics codes. “Competence in a field of historical scholarship includes . . . methodological rigor, capacity for interpretation, originality, thoroughness, and skill in writing,” we read in the 1974 AHA document. ¹⁰⁶ Moreover, we find a principle titled “competence” in every successive revision of the APA ethics code from 1959 to 1992, with “competence” even numbered ahead of “integrity” in the 1992 code. ¹⁰⁷ “The maintenance of high standards of competence is a responsibility shared by all psychologists in the interest of the public and the profession as a whole,” it was affirmed in APA documents between 1959 and 1981. ¹⁰⁸ Here, maintaining competence was particularly linked to the psychologist’s self-awareness of her own “limitations” and “effectiveness.” ¹⁰⁹ If the precise wording differs between revisions of the APA code, the repetition of verbal forms of “recognize” reveals the importance of a quality of (self-)awareness for psychologists, just like an “awareness of bias” was an important component of the quality of integrity for historians.

(Self-)awareness had a larger presence in scholarly ethics codes than as an element of competence or integrity, however, especially for psychologists. This is especially the case if we consider not only direct mention of “awareness” or “being aware,” but also the broader conceptual grouping that includes notions of “recognition,” “sensitivity,” or “sensible regard.” A psychology student should be inculcated with a “sensitivity to the ethics of the psychologist-client relationship,” that the psychologist “shows sensible regard for the social codes and moral expectations of the community in which he works,” and as a teacher, is “aware of the fact that their personal values may affect the selection and presentation of instructional materials.” ¹¹⁰ To be aware of a plurality of moral expectations or perspectives on some material is also to be open-minded, a psychological characteristic that permeated American culture during the Cold War. ¹¹¹

It is thus perhaps unsurprising that this larger sense of (self-)awareness is the principal personal trait to sit under both our umbrella qualities of integrity and competence. Otherwise, the more traditional virtues divide surprisingly cleanly between the two, across ethics codes from both the APA and AHA. Under integrity, we find honesty, fairness, open-mindedness, and promptness, while competence can be read as entailing originality, thoroughness, clarity, accuracy, and care. ¹¹² Recommendations about using good method to promote both integrity (readiness to follow sound method) and competence (methodological rigor) appear in ethical documents directed to historians, but method is hardly embodied in a researcher.

Yet if integrity and competence can be distinguished quite neatly in terms of the virtues that fall under their respective umbrellas, the corpus of ethics codes does not lead to any clear answer on how precisely to categorize integrity, competence, awareness or, additionally, “responsibility,” also a heading in many APA ethics codes. ¹¹³ Although they sometimes appear as virtues “that must be kindled,” ¹¹⁴ their ontological status becomes blurred in statements that term integrity an “issu[e] of professional conduct” for historians, declare that maintaining competence is a “responsibility,” or explain the psychologist’s “responsibility” as a matter of “plac[ing] high value on objectivity and integrity.” ¹¹⁵ This kind of ontological mixing has also been found in several quantitative investigations of “research integrity” or “values talk” in contemporary corpora that include ethics frameworks or codes of conduct among their sources. ¹¹⁶ If integrity, competence, awareness, and responsibility were important qualities for American historians and psychologists, the precise ontological status of these notions remained suspended. ¹¹⁷

Alongside these ambiguous yet important personal qualities, we encounter a certain number of virtue-qualifiers in the societies’ ethics codes, as we might expect from their importance in our other scholarly genres. Such virtue-qualifiers are similarly predominantly positive; they refer to virtues, not vices. ¹¹⁸ For example, “thorough, fair and objective evaluations” are required for peer review in physics to “serve its intended function,” ¹¹⁹ and both historians and psychologists were enjoined to “present their credentials accurately and honestly in all contexts.” ¹²⁰ Where scholars were to present something – materials to students, research to the public, or position descriptions and reports in academic recruitment – the duo of objectivity and accuracy predominate. In 1953, the APA noted that “in presenting potentially disturbing subject matter to students,” discussion should be “objective and full,” ¹²¹ a point whose virtue-qualifiers are echoed in the 1992 code’s requirement to “present psychological information accurately and with a reasonable degree of objectivity [to students].” ¹²² As well as describing any open positions “accurately,” historians selecting candidates for recruitment should ensure “fair professional practice.” ¹²³ Modesty and caution (or care) joined the constellation of virtues that qualify scholarly presentation, such as when the APA codes explained how to make public statements: “Modesty, scientific caution, and due regard for the limits of present knowledge characterize all statements of psychologists who supply information to the public.” ¹²⁴ Historians, for their part, had to “be careful not to present [their findings] in a way that foreclose[d] discussion of alternative interpretations.” ¹²⁵

Unusually for our corpus, recommendations in the 1981 APA ethics code about making public statements were constructed partly in negative terms, regarding what such statements “do not contain” – to wit, “a false, fraudulent, misleading, deceptive, or unfair statement.” ¹²⁶ If this is a rare case of negative virtue-qualifiers (vice-qualifiers?), other instances of vices in the APA or AHA ethics codes tend to be expressed in noun form and to point unequivocally to a certain behavior. Historians’ individual beliefs, urged the AHA in 1974, could not justify “the intentional use of falsification, misrepresentation, or concealment or the abuse of academic and psychic authority to intimidate students.” ¹²⁷ Together with “misrepresentation,” and “abuse” (or misuse – of authority or results), common vicious comportments in the codes included “fabrication” (of data), “discrimination” (on grounds of sex, race, etc.), and, especially from around the 1990s, “plagiarism” and “sexual harassment.”

About the same time that sexual harassment and plagiarism made their appearance as named forms of unethical behavior, some ethics codes also took on an explicit regulatory function. ¹²⁸ Notably, the 1992 revision of the APA code was divided into general principles and a number of “enforceable rules for conduct.” ¹²⁹ Since any enforcement process can only judge behavior, not who a scholar is as a person, the document accordingly specified that it only applied to “psychologists’ work-related activities,” not their “purely private conduct.” ¹³⁰ By implication, the scholarly self of a psychologist would be split off from the private self of that same individual, even as virtue talk remained prominent.

The virtues, are they a-changin’?

This significant persistence of virtue talk, taken to encompass both personal qualities and virtue-qualifiers, shows that American scientists continued to name particular virtues, even in an age when the word “virtue” itself fell out of use. Before elaborating on our findings in the conclusion, we would like to draw attention to a number of changes that can be observed in the period under discussion. We will mention three: (1) the increased prevalence of virtue-qualifiers, especially in book reviews; (2) the declining importance of “objectivity”; and (3) the rise of “integrity.”

First, there are some differences between our three genres in the balance of personal qualities to virtue-qualifiers over the decades studied. The two occur more or less equally frequently in ethics codes throughout the period under discussion, and, although virtue-qualifiers outweighed personal qualities in textbooks, the balance between them did not change significantly over time. In book reviews, however, we do observe a shift. Toward the end of the twentieth century, book reviewers in all three disciplines invoked personal qualities significantly less often, while the frequency with which they used virtue-qualifiers remained more or less constant. Even if virtues continued to be discussed by book reviewers in history, physics, and psychology alike, they were referred to increasingly indirectly, in the form of virtue-qualifiers. A possible explanation for this shift may be that postwar scientists felt increasingly uncomfortable with directly mentioning the personal qualities of their peers. We find expression of this sentiment not only in CP editor Boring’s advice to book reviewers to avoid ad hominem criticism, but also in the APA’s first ethics code (1953), in which book reviewers were enjoined to focus “on the adequacy of the work done rather than on the integrity or the ability of the author.” ¹³¹ In the 1980s, philosopher of science Larry Laudan echoed such calls when urging his fellow critics of creationism to “assess the merits or demerits of creationist theory without . . . speculat[ing] about the unsavoriness of the mental habits of creationists.” ¹³²

If book reviewers were increasingly reluctant to indicate authors’ personal qualities directly, it is important to add that they continued to regard the use of virtue-qualifiers as unproblematic. Sometimes, book reviewers were even encouraged to use this form of virtue talk. Interestingly, the AHA’s 2005 code of conduct (an amended version of which is still used) recommends those “reviewing books . . . to evaluate the honesty and reliability with which the historian uses primary and secondary source materials,” thereby linking the virtues of honesty and reliability to researchers’ practices rather than their personality. ¹³³

A second change relates to “objectivity.” As we saw in the sections on introductory textbooks and book reviews, this virtue was fundamental for both historians and psychologists in the 1950s and 1960s; however, it then declined in the 1970s and 1980s. (Recall that physicists had talked about objectivity only rarely even in the 1950s and 1960s.) In place of triumphant phrases like “objective science of psychology,” discussion of objectivity took the form of more cautious warnings against “bias,” which was increasingly presented as something of which psychologists needed to be aware. Whereas Hilgard’s 1962 textbook had instructed the aspiring psychological researcher to “remain objective in his study of the facts as they exist,” implying that objectivity was something that could be attained as a matter of course, later psychology textbooks confessed that an attitude “uncolored by personal feelings and attitudes” was so “extremely difficult” and “almost impossible” to attain that “clinical practitioners cannot be totally objective no matter how hard they try.” ¹³⁴

Historians’ attitudes toward objectivity shifted similarly. While some textbook authors tried to keep the “noble dream” of objectivity in place by arguing that objectivity should not be confused with a total absence of bias, or by granting that it could only be realized to a certain degree, others preferred to abandon the quest, embracing alternative virtues like honesty instead. ¹³⁵ Textbook authors specifically recommended historians to reveal their biases, thus allowing readers to recognize them, rather than making vain attempts to overcome them. ¹³⁶ Along these lines, a 1977 textbook quoted Gaetano Salvemini: “Impartiality is a dream and honesty a duty. We cannot be impartial, but we can be intellectually honest.” ¹³⁷ The 1987 AHA code took a similar stance by recommending “awareness of one’s own bias” as part of a historian’s “integrity in scholarship.” ¹³⁸ Historians, in other words, were increasingly summoned not to overcome their biases, but to be transparent about them – a subtle but important difference that marked the end of a long tradition of suppressing personal attitudes and preferences. ¹³⁹

Thirdly, “integrity of scholarship,” as discussed in the AHA code, points to what some have identified as a new chapter in the moral history of American science: the emergence of integrity as something between a metonym and an umbrella term for responsible conduct. While most existing literature points to the 1990s as a decade witnessing a quick rise in popularity of “integrity,” we saw that this term already featured prominently in the 1959 APA ethics code, while also occupying a privileged place in the AHA standards from 1974 onward. ¹⁴⁰ In the genre of ethics codes, therefore, the use of integrity as a virtue term significantly predates the rise of research integrity as a broad concept in academic ethics. ¹⁴¹ As noted above, however, ethics codes are ambiguous on whether integrity should be classed as a (new) virtue in itself or an umbrella quality. We should also add that our findings here refer only to codes of conduct: textbooks and book reviews did not often invoke integrity, either before or after the 1990s. This may reflect a time-lag in the pervasiveness of integrity talk (if integrity were to appear in such sources in the 2000s or 2010s). Alternatively, it may be a matter of the kinds of genres in which scholars discuss integrity (like policy documents, articles about scientific fraud, or editorials). Integrity is, after all, difficult to use in adjectival form, making it an unlikely virtue-qualifier.

Conclusion

Comparing textbooks, book reviews, and codes of ethics in history, physics, and psychology, we have observed regular use of virtue talk in all three genres, in all disciplines, over the entire postwar period in the twentieth century. Not only were certain personal attributes and forms of behavior indispensable to the pursuit of research, but historians, physicists, and psychologists continued to name qualities like carefulness, fairness, and honesty explicitly when they talked about the business of science. We can thus extend Shapin’s aforementioned thesis that “personal virtue still matters to the making and warranting” of late modern science to scientists’ discourse: talk about personal virtues also persisted among American scientists between 1945 and 2000. ¹⁴²

In addition, our research maps out some ways this discourse on virtue took on different forms. Postwar American scientists applied virtues both to individuals’ personalities and attitudes (what we call personal qualities), and also to their methods, modes of reasoning, and working habits (in the form of virtue-qualifiers). Throughout the period under discussion, these two forms of virtue talk co-existed, sometimes in entangled ways. The ratio between them, however, differed across genres. If book reviewers and textbook authors commonly made direct reference to personal qualities, their use of virtue-qualifiers was even greater, especially toward the end of the twentieth century. Ethics codes, by contrast, show a more equal balance between personal qualities and virtue-qualifiers. The most notable divergence between the three genres studied is thus a matter of how directly virtues were attached to the person of the scholar as an individual, not of which virtues they promoted. Indeed, this may well constitute a properly generic difference, linked to the textual particularities of ethics codes, as contrasted with book reviews or textbooks. Ethics codes, after all, tended to be considerably shorter than textbooks (one or two pages), while still having to cover the full range of traits or comportments that went into making a good scholar. We can speculate whether this condensed structure brought a corresponding need to name the desired qualities directly, rather than to describe them, as suggested by differences between the lengthy 1953 APA ethics code and its abbreviated successor in 1959.

Our analysis also points to several differences between fields in the prevalence of virtue talk, and in the repertoire of virtues discussed. “Objectivity” is most notable in being invoked frequently by historians and psychologists, yet largely unmentioned by physicists. That said, similarities in virtue talk across the disciplines were as important as the differences. These similarities are most obvious on the level of virtue-qualifiers. We find that historians, psychologists, and physicists mobilized the same core set of virtue-qualifiers to evaluate their peers’ work, introduce their students to their disciplines, or define their professional standards. In particular, they made frequent use of virtue-qualifiers related to “thoroughness,” “carefulness,” and “accuracy” – virtues central to what has been called an ethos of exactitude. ¹⁴³

More broadly, we have sketched out a comparative framework encompassing the constellation of virtues regularly named as relevant to doing science in postwar America, a set that is often not internally consistent, but rather comprises ideals that pull in different directions. This will open the way for future well-researched case studies that extend beyond a single author, genre, or practice. How repertoires of virtue might have articulated with dimensions of gender, race, or class merits particular attention, given that historians have already noted that a virtue like “objectivity” carried with it connotations of whiteness and masculinity. ¹⁴⁴ We already know that non-white and non-male students and researchers generally struggled to be recognized as properly scientific in Cold War America, making it highly pertinent to examine further the ways in which virtue talk was employed as a vehicle of marginalization and exclusion in this context. In addition, it seems worth comparing (academic) scientific disciplines to mixed or applied professions, like medicine, which also has a centuries-long tradition of thinking about virtues, and journalism, where notions of virtue left a deep imprint on some of the field’s earliest codes of ethics. ¹⁴⁵

For such future studies, this article offers an additional analytical resource in the novel distinction made between personal qualities and virtue-qualifiers. Indeed, in its absence, existing scholarship sometimes tends to conflate the situation of researchers embodying the quality of carefulness with research characterized by careful measurement and papers containing carefully worded arguments. Something similar happens in studies of “epistemic virtues,” especially if this category is defined so broadly as to bundle virtues like “objectivity” with what Samuel Schindler calls “theoretical virtues,” that is, properties of scientific theories like testability, coherence, and fertility. ¹⁴⁶ Our distinction between personal qualities and virtue-qualifiers could therefore fruitfully be applied to disentangling distinct elements of scholars’ virtue talk – and indeed to understanding scholarly virtues and vices in general – in earlier periods and other parts of the world. This promises to deepen our insight into the longue durée history of scholarly virtue discourse, as well as to enable us to assess what, if anything, was new about the range of virtues talked about by scientists in the postwar United States.