Mary Jo Nye: I see some new faces. [Laughter] Our next speaker is Ana Simões. Ana Simões received her PhD from the University of Maryland in History and Philosophy of Science. In fact she worked with Stephen Brush, who’s training within theoretical chemistry, and who’s a well known historian of chemistry and physics, and a real contributor in the field of science education in this country. Ana is a member of the physics department and also head of the Center for the History of Science at the University of Lisbon. She’s published a number of very important studies in the history of quantum chemistry and in the history of chemical physics. And in some of these articles that she has written, which have become absolutely necessary reading for historians and philosophers of chemistry working on twentieth century problems, she’s emphasized some of the chemical work of Linus Pauling, of John Slater, of Robert Mulliken, and of Charles Coulson. Her research interests also include the history of science in Portugal and the history of science and technology on European peripheries, and she has published of course in Portuguese as well as in English. Ana’s paper is on textbooks as manifestos, Coulson after Linus Pauling, and Robert Mulliken.

Ana Simões: Thank you very much for this introduction and thanks Mary Jo Nye and Cliff Mead, and their team for organizing this conference. So, I’ll start my talk. I will be reading it. [1:47]

In an obituary notice, the Nobel Prize winner Roald Hoffman evaluated the contributions to quantum chemistry of the trio Linus Pauling, Robert Sanderson Mulliken, and Charles Alfred Coulson which I chose to address in this talk. The Nobel Prize winner Roald Hoffman evaluated in a paper the contributions to quantum chemistry of the trio Pauling, Mulliken, and Coulson which I chose to address in this talk in the following manner:

American and British chemists had secured a place for quantum mechanics in chemistry, through the charismatic exposition of Linus Pauling, the quieter and deep reflections of Robert Mulliken, and the elegant and perceptive teaching of Charles Coulson.

Not surprisingly, he highlighted in both Pauling and Coulson their excellence as communicators while he stressed Mulliken’s qualities as a scientific thinker. [3:15]

Okay, certainly, all three were builders of the discipline of quantum chemistry, but just Pauling and Coulson excelled as teachers, textbook writers, and popularizers. In fact, in the standard history of quantum chemistry, when assessing the reasons behind the ready acceptance of the valence bond method pushed forward by Pauling, Wheland, etc. when compared to the molecular orbital method developed by Mulliken, Lennard-Jones and extended by Coulson, historians usually acknowledge that initially there were no predictive or empirical reasons to account for the early success of the valence bond method, which owed much to the persuasiveness of Pauling as a teacher and a textbook writer. Specifically, Pauling’s textbook The Nature of the Chemical Bond, which came out in 1939, was never challenged by any textbook written by Mulliken, who in fact never wrote any textbook. It was only rivaled by Coulson’s Valence, a textbook which came out more than a decade later in 1952. This was the reason behind naming Coulson’s Valence as, I quote, "the textbook Mulliken never managed to write", end of quote, an expression which I chose as the title for a joint paper’s section. [4:33]

Science is made, but science has to be communicated. Textbooks are privileged means of conveying science to students, of training them in the skills of the trade, and they often arise as the end-product of years of lecturing to successive classes of students. Viewed from this perspective, they are reified conversations between teacher and students. But textbooks have been presented as dogmatic expositions of science, stripped away of all novelty, far-away from the turmoil of science in the making, in sum, privileged means to indoctrinate students in the practices of normal science. Historians of science have recently challenged this view. They have claimed that the distinction between science in the making and science ready made is often not so clear cut, and therefore textbooks should not be seen as neutral vehicles of normal science. Following this line of reasoning, I have argued that early textbooks of quantum chemistry, and specifically The Nature of the Chemical Bond and Valence, can be seen as partisan and polemical textbooks in which authors discussed “the question of whether or not quantum chemistry is an application or use of quantum mechanics for chemical problems.” In this talk, rather than look at them as manifestos in which the status of quantum chemistry was discussed, I look at both textbooks as manifestos for particularly and distinctive epistemological views on quantum chemistry. [6:08]

In what follows I will try to explore the potential of a dialogical historiography, in which science is viewed as a result of the way scientists respond to and address each other, and scientific theorizing is endowed with a multidirectional dialogical nature, despite the fact that often in the end by means of different strategies the dialogical intercourse is flattened into a monological narrative. But I am not centering my talk on the context of discovery but rather on the so-called context of pedagogy, a distinction which is challenged in what follows. In fact the discussion underlying the writing of both textbooks under scrutiny and their successive editions was a discussion which touched upon central elements of the scientific worldviews of both authors, elements which cannot be relegated to the realm of expository science. I am therefore looking at science first and foremost as an act of communication, be it either at the production or at the circulation level, while challenging the distinction between the production of science and its various contexts of communication. And I have come to form these ideas as a result of working out the proposals of historians of science, Mara Beller and James Secord, and applying them to what Thomas Kuhn dubbed the context of pedagogy. [7:38]

Therefore, in what follows I invite you to look at Coulson’s Valence not as the textbook which ended the hegemonic reign of Pauling’s The Nature of the Chemical Bond, but as the outcome of an on-going conversation, either literal or metaphorical, between Pauling and Coulson, in which each scientist qua author was responding to the other, agreeing or disagreeing with him, in the process revising his ideas, writing his textbooks as a result of an enduring dialogue which did not stop with Valence nor even with the last edition of The Nature of the Chemical Bond. Seen from this new perspective, Valence was the textbook “neither Mulliken nor Pauling managed to write”, and this realization is at the core of contrasting Pauling and Coulson’s epistemological views on quantum chemistry. To simplify my narrative, I will center the discussion on the concept of resonance introduced by Pauling to explain the properties of molecules such as benzene for which no single valence bond structure accounts for its chemical behavior, and its associated notion of the chemical bond. I claim that resonance was a concept central to their on-going dialogue and the parallel articulation of different epistemological views on quantum chemistry. [9:00]

Having been used to associate Pauling with the articulation and lifelong defense of the valence bond method, it is instructive to point out that in the early years of his career, Pauling explored possibilities of molecular orbitals first in the framework of the old quantum theory and then in the context of the new quantum mechanics. And you have these notebooks here at the Ava Helen and Linus Pauling archives. While this early avenue was abandoned, Pauling continued to look for a quantum mechanical explanation of the chemical bond. It was finally articulated in the famous series of seven papers on “The nature of the chemical bond” in which notions such as hybridization, maximum overlapping and resonance among valence bond structures were put forward. He was then a full professor at Caltech. At about the same time, Mulliken was exploring in Chicago the same problem from a totally different perspective. [9:57]

It amounted to abandoning altogether the old notion of chemical bonds and to replace them by the alternative conception of molecular orbitals, and this was done in a long series of papers on “Electronic structure of polyatomic molecules and valence.” Despite these seemingly opposite viewpoints, which came to be known as the valence bond and molecular orbital theories, the fact of the matter was that no later than 1935 a review paper by Van Vleck and Albert Sherman proved that both approaches were mathematically equivalent. Having this proof in mind, their differences were to be found at the ideological level, a fact that Mulliken noted soon when he poignantly characterized the valence bond theory as following “the ideology of chemistry.” Mulliken was stressing the reliance of the valence bond on the notion of chemical bonds to which he objected. On this count both Mulliken and Pauling could not agree more: both accepted that Pauling’s strategy stressed the continuity of the valence bond with traditional chemical theories of structure grounded on the acceptance of chemical bonds. [11:16]

It is in this context that one should consider the failed attempt of Pauling and his former student George Wheland at writing a textbook on the Quantum Mechanics of Organic Molecules, this happening in the academic year ‘36-‘37, in which they planned to offer an extensive comparison of the valence bond and the molecular orbital methods. During this academic year they outlined its chapters and allotted tasks to each co-author. Wheland prepared the introductory chapters based on the valence bond method and Pauling was in charge of the chapters which would compare both methods. Wheland's part was soon completed and revised by Pauling. Pauling never delivered his share of the project. It is interesting to speculate on the reasons behind the failure of the only project we have evidence of in which Pauling was willing to take seriously the molecular orbital method and to contrast the performance of both methods when dealing with organic molecules. Certainly, by that time, Pauling’s attention was then drifting towards applications of larger molecules of biological interest. More importantly, this failure hints at Pauling’s inability to stand on the molecular orbital reference frame. This incapacity was certainly the outcome of the articulation of a chemical theory centered on the concept of resonance, the correlated belief that resonance stood for as real a molecular phenomenon as any other molecular property, and that resonance theory was as much a chemist’s artifact as any other theory. This much was hinted at in the first edition of The Nature of the Chemical Bond and extensively articulated in successive editions, especially in the last one, to which I will refer later. Furthermore, this was the beginning of a program of reformation of chemistry from the standpoint of the theory of resonance which was to become Pauling’s hallmark, and which was outlined in a practical way in textbooks such as General Chemistry to which Mary Jo Nye has referred previously, and College Chemistry. Quantum chemistry notwithstanding, the teaching of chemistry should get students to develop what they call a “feeling for chemistry” through the contact with chemical compounds and their properties. What you were just referring to is the different approach of chemists in relation to physicists. At the introductory level, Pauling claimed, the teaching of chemistry should not rely on extensive mathematical apparatus, but should depend on traditional structural notions such as chemical bonds. The idea of resonance was introduced as an embodiment of former structural ideas. [14:16]

For many of its readers, Pauling’s textbook pushed forward a monolithic view of quantum chemistry grounded on the theory of resonance. This is clear from the conclusion of the review offered by Robert Mayer, where he discouragingly noted that:

"It is unfortunate that this treatise will almost certainly tend to fix, even more than has already been done by the author's excellent papers, the viewpoint of most chemists on this, and only this one approach to the problem of the chemical bond. It appears likely that the [Heitler-London-Slater-Pauling] method will entirely eclipse, in the minds of chemists, the single electron molecular orbital picture, not primarily by virtue of its greater applicability or usefulness, but by the sole brilliance of its presentation." [15:11]

To reverse this situation was not an easy task. Mayer ventured that another book on the same subject, using the same examples, but emphasizing other alternative methods should help in the process, but he predicted that it would be very hard to emulate the clarity and simplicity of Pauling's almost exemplary presentation. Writing such a book was beyond Mulliken’s abilities, but it proved to be at the reach of Coulson’s. Just two years after The Nature of the Chemical Bond came out, Coulson made his first incursion at textbook writing while at the same time extending the molecular orbital method with characteristic verve. The mathematization of quantum chemistry was to become his hallmark. Away from the centre, isolated in Dundee during war time and still in the beginning of a university career, Coulson delivered a little book on waves in which he explored the unifying power of mathematics in treating wave phenomena. With missionary zeal he continued writing textbooks throughout the very end of his life. The textbook on electricity followed in ‘48, then Valence in ‘52, certainly his most renowned textbook, and finally The Shape and Structure of Molecules, one year before he died in 1973. [16:36]

At the same time, as if warming up to deliver Valence, Coulson began articulating a response to Pauling’s program outlined in The Nature of the Chemical Bond. In a paper deliberately called “Quantum Theory of the Chemical Bond,” which came out in 1941, two years after The Nature of the Chemical Bond, Coulson appropriated Pauling’s concepts of hybridization and maximum overlapping and translated them into the language of molecular orbitals. While for Mulliken there was no such thing as a chemical bond, for Coulson molecular orbital theory did not have to abandon a pictorial representation of chemical bonds. On the contrary, he filled his paper with diagrams, which later became familiar to any college chemistry student, depicting the formation of the molecular orbital in water, the formation of the “double-streamers” in ethylene, and the formation of the chemical bonds in benzene, depicted by what came to be known as the “doughnut” model. He went on to explore the notion of the chemical bond in the framework of molecular orbitals, at the same time that he became increasingly critical of the notion of resonance. For him, resonance was not definitely a molecular property but just a heuristic device, a “calculus.” These are two very important papers in 1947. [18:13]

Valence would eventually become a classic. Throughout the book, Coulson eagerly emphasized the special role played by the alliance of experimental results and chemical intuition, the feeling for chemistry Pauling was talking about, in the suggestion of particular mathematical developments: the presentation of both methods did not follow their historical order of appearance. He presented molecular orbital theory as the first topic because it was according to him, “conceptually the simplest.” Though Valence was undoubtedly a book sympathetic to the molecular orbital point of view, it did not develop any rhetoric to condemn the valence bond methods. Coulson considered that both schemata were approximations whose range of validity had been sufficiently understood for us to recognize, I quote, "the folly of trusting to either alone,” end of quote. In the book, Coulson acknowledged that resonance represented one of the most powerful ways in which chemical intuition guided one into finding suitable wave functions, and that the appeal of the valence bond method, laid in its selection of component wave functions, carry pictorial connotation. No analogous comments can be found for the molecular orbital approach. In the third part of the book, Coulson used both methods indiscriminately, contrasting them in public in a way Pauling never managed to carry out. As soon as his “little” book on valence appeared, Coulson made sure to send a copy to Pauling. In the first days of September 1952 a review of the book by Pauling appeared in Nature. [20:00]

The review was definitely hostile. Pauling felt that the treatment of quantum mechanics was brief, sketchy and many equations were derived by methods which had not been presented. Pauling, in a way, accused Coulson of over-enthusiasm for the molecular orbital method which pushed him to make various unsupported claims. Coulson immediately replied to Pauling. Two of Pauling’s remarks particularly worried him. Not unexpectedly, one concerned Coulson’s “over-enthusiasm for the molecular orbital method,” the other his treatment of hybridization, to which a whole chapter was dedicated. [20:46]

Predictably, in his reply, Pauling conceded that the chapter on “hybridization” particularly disturbed him because hybridization was a concept he “discovered (or invented),” the cornerstone of his resonance theory, and he felt that Coulson did not give him proper credit. Most of the material contained in this chapter had been first published in Pauling's 1928 paper, in which he hinted at hybridization as a possible explanation for the tetravalence of carbon. Then, it was explored in the first paper in The Nature of the Chemical Bond series, in which hybridization - not yet called as such - was discussed analytically. Despite critiques on other matters too, Pauling was willing to confess that his comments should be taken as personal opinions about questions to which there did not appear to be yet any final answers. In a subsequent letter, Pauling even explored alternative solutions to their points of contention. In a letter to Pauling, Coulson apologized for his unintentional unfairness towards some viewpoints associated with Pauling, and he confessed that, I quote, “I have learnt so much from your work myself as to make me always grateful,” and agreed with Pauling on the importance of the 1931 paper. In Pauling’s final reply, the angry tone of his former letters disappeared. Pauling may have felt that in Coulson he found a non-antagonistic and understanding interlocutor. In fact, the second edition of Valence, which appeared in 1961, incorporated most of Pauling’s comments, gave even more weight to his two earlier papers, and further emphasized resonance. Though Coulson had declared impartiality as to being on the side of the valence bond method or the molecular orbital method, the fact remains that the latter viewpoint received an impressive boost with Coulson’s Valence. [22:57]

On the other hand, Coulson was uncharacteristically assertive about the status of resonance. He objected to Pauling’s choice to raise resonance into a chemical category. For Coulson, against Pauling, resonance was just a heuristic device, an algorithm or a metaphor or simply one pedagogically expedient method - out of many - for understanding quantum chemistry. Pauling advocated the opposite viewpoint, and he deemed the topic so important that he made his position public in Perspectives in Organic Chemistry and later on, in the third edition of The Nature of the Chemical Bond, which came out in 1967. More than the question of the artificiality of the resonance concept, to which he alluded briefly in his Nobel lecture, he wanted, once and for all, to state as clearly as possible his views on theory building. In the preface to the 1967 edition, Pauling pointed out that the theory of resonance involves, and I quote, “the same amounts of idealization and arbitrariness as the classical valence-bond theory,” end of quote. A whole section was added to discuss this question bearing the revealing name, “The Nature of the Theory of Resonance.” [24:30]

When looking at the different phases The Nature of the Chemical Bond and Valence went through as a result of their authors’ on-going dialogue, one is struck by the evenness of the two discussants: both were charismatic leaders and enthralling teachers, both shared humanitarian commitments and the appeal for popularization. Despite being usually presented as scientists on opposite sides of the chemical bond fence, the fact of the matter was that they were in agreement on many counts. They agreed on the importance of what counts as “appropriate” or “effective” training namely, on the importance of quantum mechanics and mathematics for chemistry. They agreed also on the possibility of avoiding sophisticated mathematics at the introductory teaching level. They agreed on the development of conventions for the moderation of technical jargon, including the recourse to visual representations as substitutes for or complements to advanced mathematics. They were both keen on exploring the changing boundaries between chemistry, physics, mathematics and biology. They were both sensitive to the role of audiences and they both acknowledged the importance of persuading organic chemists by appealing to traditional chemical modes of thoughts. But they could not agree on everything. [25:56]

They disagreed on the role of a comparative methodology, on the advocacy of realism versus instrumentalism, as well as on the ontological status of resonance and generally of scientific concepts in shaping science. This may explain why, as late as 1970, Coulson could assert that “resonance is a dirty word” and that ten years later Pauling disparagingly commented that, quote, “it was a tragedy that the writers of elementary textbooks of chemistry decided to discuss the molecular orbital method…” end of quote. How can we explain such late antagonistic claims? Precisely because behind their divergent views on the use of the comparative method, and on the meaning of resonance, lay their divergent views on the epistemological foundations of quantum chemistry, and generally of science, which their dialogical intercourse helped to articulate. I contrast them by naming Pauling’s approach to science as “monolithic” and Coulson’s approach as “ecumenical.”

What I mean is that the publication of The Nature of the Chemical Bond has to be assessed in the context of an ambitious strategy articulated by its author and pursued by general chemistry and college chemistry. Together, with his other textbooks, Pauling wished to implement an agenda aimed at nothing less than reforming the science of chemistry from the standpoint of quantum chemistry. This agenda had far reaching implications in what concerned the status of chemistry within the hierarchy of the sciences. Believing in what he called “integration” of the sciences, which he deemed to be achieved through the transfer of tools and methods, Pauling elected what he called the “technique of thinking” as the most important kind of transfer. It is in this respect that he came to view chemistry, and specifically resonance theory, as playing a pivotal role within the physical and biological sciences. For Pauling, chemistry played a central part, one formerly played by physics. He believed science as a whole should be reformed from the standpoint of chemistry. No more, no less. [28:21]

On the other hand, the importance ascribed by Coulson to the application of the comparative method made him particularly receptive to the potential of dialogical reasoning. In a sense, the belief in the power of the comparative method reflected his views about quantum chemistry. He was a firm supporter of the view that, “the validity of the scientist’s account depends on the degree of interlocking between its elements." To elaborate on what he means, he appropriated Charles Peirce’s cable metaphor, according to which, “the strength of an artificial fiber depends on the degree of cross-linking between the different chains of individual atoms.” In an analogous manner, the explanatory success of quantum chemistry rested on the degree of interlocking among constitutive elements – its different subcultures with their varying methodological allegiances, its different communities sharing values, commitments and approaches – to such an extent that it was not the relative contribution of each component that mattered, but the way in which the whole was reinforced by the cross-linking and the cross-fertilization of all its elements. [29:40]

I have analyzed how both The Nature of the Chemical Bond and Valence evolved in time, by looking at the context in which they both arose, the modifications they went through in successive editions and how these modifications could be correlated with the successive stages the other textbook was concomitantly going through. I have also correlated their evolution with other relevant outputs by both authors - review papers, correspondence, manuscripts - or even outputs by authors such as George Wheland. From this perspective, both textbooks are seen not just as reified conversations between teacher and students but also as reified conversations among both authors who in the process sharpened their views about quantum chemistry and afterwards conveyed them to readers and students. As chemistry has often been claimed to be a science in which conversations play a prominent role, this new vantage point looks at textbooks as conversational devices, in many respects at the same level as conversations central to the forging of science itself. And I thank you. [Applause] [31:00]

Mary Jo Nye: Thank you very much Ana. Do we have a couple of questions or comments? Yes.

Audience Member: Your mention of conversations, your emphasis on communication…

Ana Simões: Yes.

Audience Member: Your reference to missionary zeal caused me to think, whether in contemporary science do you think there are pressures to make scientists somewhat less propagandistic. Make them less…they often seem like the communications and conversations are advertisements either for themselves or for their specific field, or for the science as a whole. Do you see what I mean? That this would seem to interfere, I would think, with the kind of conversation that you obviously favor?

Ana Simões: I think that, for these two people, which were pushing forward a new discipline with very precise values and commitments, of course it was the case that they wanted to present to the public in a very precise, and also nice way, the building blocks of their theory. They were also conversing among each other. As to your question, what I think is that nowadays there is an immense pressure for scientists to speak out their ideas for as large audiences as possible. Of course, this sort of process is a process in which you always have to push forward your ideas in a very affirmative way. There is still an idea of missionary zeal for many reasons. It can be because you have to prove to your financial agencies that your science is worth being financed. It’s not always just for enlightening the audiences, but in any case, I think that this is something that is going on more forcefully now than in the time of Pauling and Coulson, in which period there were innocent singular cases. There were not many scientists. Mulliken could go on forever without having to relate to any lay audiences. He was not able to do this for personal reasons and he could do just his own research. And now I think the time has come for scientists to work on many different realms and especially this popularization of science. [34:40]

Mary Jo Nye: Robert.

Robert Anderson: I just want to make the point that I went to Coulson’s lectures in, I think, 1964 when I was studying chemistry. As far as I can remember, he presented his views very straightforwardly without actually going into any question of intention with Pauling’s ideas. In fact, I remember a reference to Pauling, but it was a long time ago now so…

Mary Jo Nye: I just thinking when Ana was talking about a story that Coulson told about when he was first appointed at Oxford in theoretical physics, that when he came to his office for the first time the sign on his door, instead of saying, “theoretical physics,” said, “theatrical physics.” [Laughter] And he said it really was a new field in those days. [Laughter]

Ana Simões: The journalists usually refer to him as the chairman of theological physics because of his Methodist commitment. There was some confusion about what really is theoretical physics. [35:55]

Mary Jo Nye: Dudley, do you have a question?

Dudley Herschbach: Yeah, I was raised, in terms of education, more valence bond, Pauling’s picture. Then, later in life, I discovered molecular orbitals and found it far more convenient. And that seems to be the general phenomenon. Few people use molecular orbitals… the valence bond compared to molecular orbitals now in everyday conversation with chemists. And I think the reason, though, is worth pointing out. Because of the force of Pauling's personality, he was good at promoting valence bonds. It was an error when he and others were using x-ray structures to electron diffraction, so they were determining what we think of in textbook terms as structure. Atoms are connected this way and this way. That’s very natural to discuss in terms of the valence bond picture, but after World War II you had spectrometers available.

Ana Simões: Of course. [37:06]

Dudley Herschbach: And chemists began looking at a lot of excited electronic states. It’s much more natural to discuss those in terms of molecular orbitals…

Ana Simões: Of course.

Dudley Herschbach: …than valence bonds. That’s really the fundamental reason why we have this total shift and now molecular orbitals are much more in the language that everyday working chemists actually use. The experimental tools have an awful lot to do with the theoretical vocabulary that’s adopted at any given time. [37:38]

Mary Jo Nye: One more question, then we’ll have a coffee break.

Audience Member: Just a contribution to the discussion, I once at an introductory chemistry conference in 1970, heard Linus Pauling expand on the viewpoint that molecular orbitals have no place in introductory chemistry.

Ana Simões: He was saying the same thing in 1980. Despite the fact that, for any practicing chemist, this change was smooth because molecular orbitals were more convenient when dealing with excited states, the fact of the matter was that Pauling was very stubborn in his advocacy of valence bonds till the very end.

Dudley Herschbach: I never asked him, but I wonder what he thought of the Woodward-Hoffman rules, which are very big in our area. They are stated entirely in terms of molecular orbitals.

Mary Jo Nye: Ana, thank you very much, and thank you all. [38:39]