Mary Jo Nye: So I would now like to introduce our first speaker, Michael Gordin. Michael Gordin is an associate professor in the History Department at Princeton. He teaches the history of science and Russian history. He is the author of a wonderful book on Mendeleev called A Well-Ordered Thing: Dmitri Mendeleev and the Shadow of the Periodic Table. He also has just recently written and released a book of something entirely different: Five Days in August: How World War II Became a Nuclear War. Michael is currently working on an international history of the development of the first Soviet nuclear bomb in 1949, and he is also working on a book on the development of nationalism in Russian and German chemistry in the mid-19th century. He is spending the academic year in Berlin, at the Max Planck Institute for the History of Science; so although he teaches at Princeton, he’s jet-lagged like our other international speakers. Michael. [Applause] [1:28]

Michael Gordin: Can everybody hear me? Thank you very much Mary Jo for the generous introduction, and thank you for bringing me out here. It’s a pleasure to be here for something related to Linus Pauling, whom I’ve always admired very much even though my work does not directly concern him. Don’t hold that against me. Despite the title of this paper, “Periodicity, Priority, and Pedagogy,” I’m not actually going to talk about the second term. Almost every time someone talks about the periodic table, they talk about who discovered it first, and give credit to someone. And, I actually have no idea who discovered the periodic table of chemical elements, and I’m going to tell you why. When you open almost any chemistry textbook you will find a sidebar next to one of the many periodic tables in it, showing some grizzled, bearded guy and calling that person the “discoverer of the Periodic Table.” That person is almost always Dmitri Ivanovich Mendeleev, who died a century ago this year, a St. Petersburg chemist of some notoriety. Sometimes he shares that space with Julius Lothar Meyer, who is a German chemist; a rough contemporary of Mendeleev, about five years older than he was. About a century ago, German textbooks might have just shown Meyer, a couple of esoteric textbooks would have shown a guy named John Newlands. All of these guys have grizzled beards and look roughly similar, and they’re all given credit for the discovery of the periodic system. [3:02]

The textbooks in chemistry today, when they do their history, have a certainty that I actually don’t possess. They know what the periodic system of chemical elements is, and therefore they can tell you who discovered it first. I actually don’t. What is the periodic system? When you talk about a discovery and try to give credit for it, usually you have to pick what you think the essence of the discovery is, and then see who crossed that particular finish line first. The problem with the periodic system is there is no consensus about what “the thing” is. Is it the recognition that there’s a periodic variation of properties of elements when they rise with atomic weight? Is it representing that particular regularity in a two dimensional grid? Is it classifying all the known elements as opposed to a subset, leaving gaps for elements that should exist but haven’t yet been discovered? Is it correcting empirical values, incorrectly - usually atomic weights - that have been incorrectly determined based on where they should fall in your two-dimensional grid? That’s called retro-diction, usually used by philosophers of chemistry. Or predicting properties of new elements? And people generally find one of these as the crucial thing, and then they find the person who discovered it first. So, who was first? Well, if you believe category one or category two, it’s probably Alexandre-Émile Béguyer de Chancourtois who discovered this thing around 1862-63, which is a layering of the elements along a cylinder, which he called the “telluric screw,” by atomic weights, and the properties go down vertically like this. Or, it could be this guy, John Newlands, who discovered a law of octaves in 1864, which is arraying all the elements by atomic weight and finds that after seven elements you start getting a repetition. It’s seven, instead of eight, because there aren’t noble gases for about three decades. Or, it could be William Odling, also British. It could be Gustav Hinrichs who was American, but Danish in origin. Lothar Meyer, or Dmitri Mendeleev. If you tell me who you think discovered the periodic system, I will tell you what you think the periodic system is. This is an amusing philosophical parlor game, we can play it later, but it’s very bad history. And it’s bad history because it presumes our current understanding of the periodic system and projects it backwards in time to see who did what when. [5:28]

So, what’s a better way of doing this? Why did all this happen from 1862 to 1869? Mendeleev discovered his table in either 1869 or 1871, depending how you count. Meyer discovered his in 1864, 1868, 1870, 1873, depending how you count. But these are all happening roughly in the 1860s, and the question is, why? Why are all of these things happening quasi-independently in one decade? And the answer has a lot to do with pedagogy. In September 1860, a congress is convened by August Kekule’—who is the person who discovered the structure of the benzene ring, bringing us back to Pauling again—to discuss how we could reform pedagogy in organic chemistry. Because of recent developments in the science, he felt that there was no standardization and thus a lot of confusion. And he wanted to bring as many international chemists as possible together in the sleepy, southern German town of Karlsruhe, to discuss ways of standardizing the practices of chemical teaching. At this conference, two chemists, Dmitri Mendeleev and Lothar Meyer, who were very young and in attendance, both vividly remember a speech given by this man: the Italian chemist Stanislao Cannizzaro, who proposed reviving the 1811 hypothesis of Amedeo Avogadro to standardize atomic weights. Of course he called it Avogadro’s hypothesis, but Mendeleev called it Charles Gerhardt’s hypothesis because priority has a lot of problems in this period. [7:04]

Once that happens, once you get a standardization of atomic weights which lets you do things – like realize that Carbon doesn’t weigh 6 but 12, Oxygen doesn’t weigh 8 but 16 - you can start laying them in order and seeing regularities you couldn’t see before with the heterogeneous systems of atomic weight determination. And, once the atomic weights get standardized, periodic systems pop up on two continents in a huge variety of different countries basically independently. And the 1860s, since it was a period when basically every concept was up for re-articulation, re-definition, or rejection, makes a very interesting pedagogical moment in how people should write textbooks. There are six periodic systems in one decade, none earlier.

So what does this have to do with pedagogy? I’m going to take two textbooks very seriously. The first is Dmitri Mendeleev’s Principles of Chemistry (that’s the Russian title), which first appears in 1869 to 1871 in two volumes, and is revised through eight editions through his lifetime. The second is Lothar Meyer’s Modern Theories of Chemistry, which appears in 1864 and is revised six times during his life before he dies in 1895. By exploring the periodic system, I want to see how it fits into the composition of these textbooks. I don’t want to ask who discovered what first. I want to ask what did that have to do with it? What did they think this system was supposed to do in the construction of a pedagogical textbook? And, in both cases we’re going to see that what they really wanted to do with the textbook, and with the periodic system, is define what they thought chemistry was supposed to be about. It was almost a boundary marker of where chemistry ends and where physics begins, and as a result it structured their entire vision of how introductory chemistry should be presented. I’m going to conclude with a sort of vexed question which is, not so much why this guy decided to predict new elements, but why this guy refused to. And that’s why he no longer gets any credit, because he didn’t do something. [9:12]

Now we shall discuss Mendeleev. His pedagogical origins, and his pedagogical jobs later, are all very heavily confined in St. Petersburg, the capital of the Russian empire in the 19th century. He’s born in Tobolsk, Siberia which is quite a ways away, but he entered as an undergraduate at the chief pedagogical institute in St. Petersburg in 1850, which was his father’s alma mater. He graduates in 1854, gets a masters in 1856, teaches at a general gymnasium in Odessa for two years, which he hated because he couldn’t stand teaching high school students, and then he gets a two to three year renewable post-doctorate. It ends up being two years, and he moves to Heidelberg where he is supposed to go and study with Hermann von Helmholtz, Robert Wilhelm Bunsen, and Gustav Kirchhoff. He actually doesn’t study with them. He goes there, he sets up his own apartment, he puts a lab in it, and he researches laws of cohesion through the capillarity of organic liquids.. He talks only to Russians who are also post-doctorates. The only German he seems to have time for is Emil Erlenmeyer. While he is at Heidelberg, though, he does happen to take a short trip to Karlsruhe and attends the conference, which he claims had a huge impact on him at the time, and it certainly did when he returned to St. Petersburg. He comes back in February 1861, two weeks before the serfs are emancipated, and he is basically starving and looking for work. He gets an adjunct position at St. Petersburg University in the fall of 1861. Within six weeks, the University is shut down for two years because of student unrest and he is jobless again. He decides to write a textbook as a way of earning money quickly. He writes one on organic chemistry very quickly, releasing it in late 1861. It wins the Demidov Prize from the Academy of Sciences in 1862, thus allowing him to get married because he now has some money, and he ends up getting a job on the strength of his textbook at the St. Petersburg Technological Institute in 1864. He doesn’t like that job very much; it’s teaching engineers. He would rather teach introductory general chemistry, preferably inorganic chemistry,– so he gets a job in 1867 at St. Petersburg University teaching introductory inorganic chemistry. This is actually a great time to have this job. In 1861 the emancipation of the serfs initiates a series of massive reforms that restructure Russian education and society in general. There’s a tremendous rise of interest in the natural sciences and also a tremendous expansion of the university population. In turn, the natural sciences faculty where Mendeleev teaches has a huge boost of students. Those students want to study various sciences, but every first-year student in the faculty has to take introductory chemistry. Because of this policy, he has a huge population of students to educate. He needs to use a textbook for this, and he has basically two options. He can either take an English, French or German textbook, translate it, update it, and give it to them, or he can write one from scratch. Russian inorganic chemistry textbooks aren't floating around at the time. He decides against the translation option because it takes a long time and the book will probably be outdated given the tremendous transformations of the 1860s, and he’ll get better royalties if he writes it himself. So he writes it himself. He starts in 1868, and spends the whole first part of 1868 and January of 1869 composing Volume 1. Volume 1 is very interesting. It basically deals with four elements for 500 pages: hydrogen, oxygen, nitrogen and carbon, the organogens. He spends about a hundred pages or so on each element, and it’s a very praxis-oriented book. He talks about oxygen: how do you find ozone, how do you purify it, how can you separate out 02 from it, where can you find 02 in nature, how do you make it, how do you isolate it, what reactions does it go through, etc. It gives you a very multivalent lab-based idea of the properties of each of these elements. And then at the end he tosses in the four halogens because that’s a clearly recognized family. [13:16]

That’s all well and good, it takes about 600 pages to do that, and he leaves 55 other elements for the 600 pages of Volume 2, and he clearly can’t do this. At least, no one would want to read that book. So in early 1869 he has a tremendous problem of how he’s going to jam 7/8 of the known elements into the same space in which he dealt with 8. This is a problem and he sits around for a month and a half not knowing what to do, trying to group them together in some way so that he can deal with a group as one, much like he did with the halogens. In February 1869, he develops the first periodic system of Mendeleev. It’s the last of all of those six guys I mentioned earlier. It’s peculiar in lots of ways. First it increases from top to bottom. It also has strange features: the halogens are over here, the alkali metals are over here right next to them. That’s how he started the table, he put them together and then he built outward from that point. Nowadays this would be at the extreme left, this would be at the extreme right, well there would also be noble gases but that’s much later. Other peculiar features of this are that it has, sort of, unfinished parts in it - this doesn’t belong here but he doesn’t know where to put it - and it has these question marks. These are gaps. People have left gaps in systems of elements before. What’s peculiar is that he starts giving them atomic weights, predicting what they should be, where they belong in the system. And he spends the next couple of years revising this thing. He does a lot of things with the system, not only giving more detailed predictions of these things, but he also corrects uranium which should be around 238, so he doubles it, and so on, refining this system until it eventually looks like this in 1871. This is the one that’s in his textbook, and it’s very much a pedagogical system. It has lots of space, you can write in stuff when new elements are discovered. It has lots of properties written down. It folds out from the book so that when you’re reading you can actually just look at this other one that’s at the edge for a general guide. It has lots of bonding properties up here. It’s a very interesting, very developed periodic system built into his book. I could talk a great length about that. [15:41]

What’s strange about his table is what he says about it in the textbook. He actually doesn’t expand on its properties very much. When Mendeleev sends a research article about it to Emil Erlenmeyer who’s the editor of Liebigs Annalen in 1871, he says: “I want only that you will pay attention to the fact that I do not set up any hypotheses because in my view these often seduce students as false keys and thus tend to slow down the free development of science.” That was, of course, a direct quote from Newton. For Mendeleev, this thing has no hypotheses associated with it at all. It’s just an almost theory-free arrangement of elements that can be taught to students without delusion, confusion, or distraction. That’s important because Mendeleev doesn’t believe in a lot of things people believe in, like atoms or valency. He doesn’t believe in Prout’s hypothesis, which originated as the idea that all elements are composites of hydrogen atoms, but later became short-hand for the idea that elements have substructure. He doesn’t believe in the electron in 1897, he doesn’t believe in the noble gases in the 1890s, he doesn’t believe in radioactivity when it first happens either. Mendeleev resists basically every single conceptual innovation in chemistry from the moment he makes the table, especially in the textbook. He argues that these are not the kinds of things you should teach students about. He’s fundamentally, both politically and pedagogically, very conservative. Except for one thing, which is this: he predicts three new elements. This is something no one has ever done before. He predicts the detailed properties of three elements based entirely on the structure of the periodic system: eka-aluminium, eka-boron, and eka-silicon. He figures by the time he dies maybe someone will discover one of them. He publishes the detailed predictions in 1871. Within 15 years all three of them are discovered. Mendeleev becomes hugely famous in the chemical world and this is the basis of his reputation because it is the one thing where he uses theoretical speculation ahead of the pack of chemists. This is the thing that most historians of chemistry and most chemists pay attention to, but it’s peculiarly unusual for him. He also turns this into a theory of why he should get credit, even though he has produced the last periodic table, because he’s fully developed the system. He is the scientist who’s really put the nail in the coffin and he makes prediction the essence of discovery. [17:59]

Meyer is very different from Mendeleev in a number of ways. He’s the son of a doctor and his mother is from a medical family as well. He was going to be a doctor, and when he is fourteen he gets pulled out of gymnasium due to a lot of headaches, and therefore ends up entering university in 1850 - the same exact time as Mendeleev. He’s not a local person. He moves around all across the German-speaking world studying with various different mentors. Mendeleev doesn’t really like mentors but Meyer does. These are the places he goes, and these are the people he studies with. He starts out as a medical person in Zurich, and then he moves to start looking at medical chemistry. Eventually he goes to Heidelberg where he loves Bunsen and gets more and more interested in physical chemistry, and finally he moves to Königsberg, which is now Kaliningrad, in Russia to study physics, where his brother happens to be as well. So he moves from medical chemistry to physical chemistry, and what unifies this entire process is an interest in chemical theory, which he thinks is the only way to unify these various spans of an extremely broad science. While he’s at Heidelberg he does stop by the Karlsruhe Congress, and makes this wonderful notation later when he edits the speech in German about the scales falling from his eyes, when he heard Cannizzaro’s speech. He gets a huge number of jobs as well while moving around. He’s an adjunct or a Privatdozent in Breslau, which is now Wroclaw in Poland, then he gets his first position at a forestry school, and then from there to a technical school, and then finally to Tübingen where he stays for twenty years until his death. He trains a very large number of graduate students while Mendeleev only trained two. Mendeleev taught undergraduates almost exclusively but Meyer really enjoys the process of raising students through graduate school. He also has a very broad, technical teaching process experience. which emphasizes his interest in theory. [20:04]

So, in 1864 when he’s still quite young, he publishes Modern Theories of Chemistry. At about 167 pages, it's not very long, certainly not yet a textbook. Instead, it’s more of an introductory treatise on the value of chemical theories. It’s built in two parts around two theories, arguing that you can explain all of chemistry if you understand atomism and if you understand valency. Atomism gets you through the structure of matter, valency talks about how bonding works. He is able to get inorganic and organic chemistry together in 164 pages in this slim, slim volume. In the 1864 edition he produces this: this is a periodic table, basically. It doesn’t have all of the elements in it, but it has here the valencies up at the top, and here it has the differences in atomic weights between different rows going down. He produces this thing, and I’ll tell you what he says about it in a minute, but he puts it in the middle of the book as a unification of the two theoretical strands of the book: atomism and valency. In 1872 he produces a second much more elaborate and accurate edition after Mendeleev’s table , which he cites.. It has Mendeleev’s predicted elements in it, it has more of a slant, kind of like that first table I showed you, and it has some structural differences from how we now do things but it’s very clearly a periodic system. And he has a lot to say about it. In articles that he publishes for research journals he also includes a graph from his text book. Okay, here, that is the elements laid out by atomic weight and this is a measure of their atomic volumes. And this curve has a clear mathematically periodic structure to it. Today, he is usually associated with mathematizing and showing how physical properties have moved along with the periodic system. He also made a periodic table. In fact, he made table five years earlier than Mendeleev made his. What does he say about these things in the first textbook? This is what comes right after he puts the table forward: “It is surely not to be doubted that a definite regularity,” he uses the same term Mendeleev uses for “law” so it’s hard to translate, “prevails in the numerical values of atomic weights. It is rather improbable that it is as simple as it appears if one leaves aside the relatively small deviations and the values of evident differences. In part, indeed, these deviations can justifiably be seen as brought about through incorrectly determined values of atomic weights. But this can hardly be the case for all of them, and entirely certainly one is not justified as, is seen all too often, due to a suspected regularity to want to arbitrarily correct and change the empirically determined atomic weights before experiment has set a more exact determined value in its place.” So right after synthesizing valency and atomism into one big beautiful picture, he specifically says you could predict things from this, but you shouldn’t because it’s not good chemistry. Later on there’s lots of quotations from the book about what chemistry’s domain is and what physics’ domain is. Physics is about prediction; they have very good laws so they know what’s going on. Chemistry is just starting to get those laws, and you shouldn’t get ahead of yourself and predict things based on a possible regularity. He specifically excludes from 1864 onward the idea of prediction. He notes Mendeleev does it, but he doesn’t think it’s a good idea --which is interesting because he is basically radical on everything else. In Modern Theories from the first edition onwards he endorses atomism, valency, a modification of the substructure of atoms and structure theory in organic chemistry. All of which Mendeleev rejected, and all of which we now think are correct. Basically, on every single theoretical issue of consequence, except for prediction, Meyer is right and Mendeleev is wrong. There’s clearly a principle at stake that he does not want to engage in prediction. Otherwise, he’s perfectly happy to teach students all of these theories that Mendeleev thinks are too sketchy and speculative. His theory of credit is also very different. He thinks it’s all about continuity and citation and gradual evolution over time while Mendeleev emphasizes radical, romantic, genius ruptures. [24:16]

Okay, what do we make of this? In analyses at the time, when they both get the Davy Medal and so on, in 1883 I believe that is, Mendeleev is always characterized as bold. In Russian, English and German you see that same word, “bold” over and over again. Predictions are bold and daring. Meyer, however, is somehow timid because he was at the precipice but refused to jump. He was too scared. Now, Meyer’s pedagogical and theoretical radicalism is quite apparent. He’s quite scared of doing a lot of things. He just doesn’t want to predict. So why do we have this case that Mendeleev is conservative on theory yet still insists on predicting, and Meyer’s radical on everything but thinks prediction shouldn’t happen? I think it has a lot to do with disciplinary borders between chemistry and physics. Both of them believe that chemistry and physics are different domains. They don’t think the boundary between them lies where we would put it today, but they believe there is a boundary of some sort. Meyer thinks the purpose of a chemistry textbook is to teach people theories that are useful as chemical theories. And he writes his articles for professional chemists and his textbooks in the same language with the same set of theories. There’s no difference between training and practicing. You should be exposed to these theories from the beginning and learn how to grow up thinking like a chemist. Thinking chemists don’t do predictions. They do research and they do empirical measurements. [25:47]

Mendeleev on the other hand, thinks that you shouldn’t use curse words around the children. You shouldn’t use theories in a textbook because it’s not appropriate. They’re not old enough to know how to handle these dangerous things so you teach them praxis and non-dangerous stuff instead. And then in his articles, he’ll do predictions for professional chemists because they know how to handle these particularly dangerous weapons. But he also thinks it’s something that physicists do and Mendeleev wants to be a physicist. He spends the 1870’s doing experimental gas research that is entirely in the domain of physics and the predictions are his leap out of boring, plodding chemistry into daring physics. The difference between the two of them is what they think belongs in a textbook, what belongs in a journal, and what is a physicist’s job and what is a chemist’s job. This is why it is important to take a look at the textbook origins and not just the articles. If you just look at the articles, you’re mystified by why Meyer won’t do predictions and you don’t think that Mendeleev is conservative on a whole bunch of other issues. But, if you look at the pedagogic roots you start seeing that these textbooks are designed to do something. Fundamentally, they are designed to train students. By looking at how they want to raise a chemist, you get a sense of what they think a chemist is. Both of these books, even though they’re rather out of date, have quite a bit to teach us. Thank you very much. [Applause] [27:21]

Mary Jo Nye: Thank you very, very much. That dovetails nicely with the theme about Pauling and Pauling’s theoretical approach.

Michael Gordin: I hope so. [Laughs]

Mary Jo Nye: Yes. [Laughs]

Michael Gordin: [Laughs] Being in the 19th century and on a different continent, I thought I should do something.

Mary Jo Nye: We have time for a couple of questions. Any questions? Okay. [27:43]

Audience Member: Was there a prominent scientist at the time who opposed the idea of a boundary between chemistry and physics? Somebody who counts? There must have been somebody who makes a difference?

Michael Gordin: There are lots of people; there are lots of arguments about where that boundary is, especially when you have collaborations like Bunsen and Kirchhoff… I’m bringing coals to Newcastle because Mary Jo knows more about this than I do. There are lots of collaborations between chemists and physicists where they are negotiating that particular boundary. Everybody seems to believe that boundary exists, that there’s not just one science. There are certain kinds of questions that physicists ask, such as related to the kinetic theory of gases. They do the kind of work where they don’t care what stuff is made of and they just want to know how it behaves. Chemists want to know what it’s made of. So when you get people like Ostwald, who tried to do thermodynamic applications onto chemical reactions, that’s when you start to see people objecting and saying "well, that’s not chemistry, or that’s not physics." So you do have certain people, Ostwald probably being the primary one, who really push that boundary and argue that they can transition it. But, they still think it’s there. I don’t think there’s anybody who thinks there’s just one science of the physical world. They understand that there are different practices and, particularly if you work in organic chemistry, you really believe there’s a difference because it’s really important what stuff is made from and how you got it. You can’t just exclude those by generalizing onto a mathematical framework. [Pauses] But Mary Jo wrote a book in 1993 that’s very good on all of this. [29:25]

Dudley Herschbach: I’ve always told students that physicists like to discern general principles that are independent of such details of what the substance is, or really like. And chemists want to know why one substance behaves differently than another. So then I say it’s a matter of epistemology, a fancy word but, I tell physicists that they have to understand that chemistry is like an impressionistic painting. You stand too close to such a painting and it’s meaningless dabs of paint, if you stand too far away it’s a meaningless blur, but at the right distance you see a wonderful thing. The physicist who wants to stand too close because he wants to elucidate the first principles, sees just the paint and says the chemist is blocky. The old fashioned biologist doesn’t want all this detail. Now, of course, the biologists are all chemists. [Laughter] [30:20]

Michael Gordin: I think that’s exactly right. Mendeleev’s tack is in a sense easier than Meyer’s, because Meyer wants to say you can speculate but only so far and no further. Whereas Mendeleev thinks that at this stage you have to stand this far from the painting, but eventually you want to get up close. Meyer thinks you never want to get that close. You kind of want to hang back at a certain distance, but you need to know where to stop. Those textbooks are very interesting because he constantly does this dance and he doesn’t know where, or when, to end. But that painting analogy is really good. I should use that!

Mary Jo Nye: One more? Okay, thank you so much.

Michael Gordin: Thank you all very much. [Applause] [31:02]