The Philosopher’s Quest

A recent book tries to carve out existential independence for chemistry.

By Philip Ball | September 17, 2017
Science History Institute

Eric Scerri and Grant Fisher, eds. Essays in the Philosophy of Chemistry. Oxford University Press, 2016. 424 pp. $60.

Some sciences are said to suffer from “physics envy”: a desire to give themselves at least the appearance of a mathematical rigor and exactitude to rival that of physics. But chemistry seems to suffer from physics anxiety: a fear it will be dismissed as ultimately “just physics.”

The argument often invoked by triumphant physicists is that all the basic components and processes of chemistry can be explained by quantum physics. The work of Niels Bohr, Erwin Schrödinger, Wolfgang Pauli, and others showed that quantum mechanics can explain why the iconic periodic table (what Eric Scerri, who has long championed the notion of a philosophy of chemistry, calls “easily the biggest idea in modern chemistry” and arguably its central “law”) has the famous twin-towered shape it does. Bohr, as well as Walter Heitler and Fritz London, also furnished the fundamentals of a quantum theory explaining how atoms join together in particular configurations to form molecules and those molecules in turn form larger aggregates, such as liquids and solids. Thanks to quantum physics, which describes how electrons (the raw currency of chemistry) are configured around atomic nuclei, it seems possible to construct chemistry from the ground up.

But does this physics-based account really “explain” all the principles that chemists use? Defensiveness against the encroaching hegemony of physics is a recurrent theme in this collection of essays edited by Scerri and fellow philosopher of chemistry Grant Fisher, and their authors offer several arguments for why chemistry’s central concepts are in fact unique to the discipline and not just an outgrowth of physics. But the case for rejecting the latter idea is broader than these essays imply. In some ways the arguments offered here even risk playing along with the game of reducing chemistry to the physics of atoms and their inter-actions. I’ll come back to this.

The notion that chemistry has a philosophy at all is relatively new, which might seem odd because the idea of a “chemical philosophy” considerably predates the emergence of physics as a discipline in the 19th century. Historian of science Allen Debus was one of the first to demonstrate that some natural philosophers in the late Renaissance saw chemistry as the key to understanding the relationship of nature and the cosmos to humankind. The Swiss physician Paracelsus (1493–1541) proposed that natural processes mirror those that could be conducted in the alchemical laboratory. Rain, for example, was the analog of condensation of vapor in a flask, and even the biblical creation of the cosmos was a divine act of chemical separation. The human body, Paracelsus said, works on chemical principles, which is why it can be treated with specific chemical remedies. This view never went away in medicine, but as a philosophy of the wider world it ultimately could not compete with the quantitative predictive power of Newtonian mechanics.

From an electron’s point of view the concept of a person is meaningless. Does that mean the concept really is meaningless?

The latter—now seen as a central pillar of early physics—came to represent the dominant model of the entire scientific enterprise, leading Immanuel Kant in the 18th century to proclaim chemistry not yet a science because it lacked a mathematical formulation. Throughout much of the 20th century, discourse on the so-called Scientific Revolution was typically hung on the framework of mathematical mechanics and astronomy. Not only did this sideline the breadth of interests of Galileo, Newton, Descartes, and their peers, including pioneer of early chemistry Robert Boyle. It also neglected to mention that Newtonian mechanics said nothing—and continues to say very little—about vast swathes of science, from medicine to zoology, botany, earth sciences, and (yes) much of chemistry.

The hegemony of physics extended also to the philosophy of science. In his 2004 book What Makes Biology Unique? evolutionary biologist Ernst Mayr rightly complained that almost all philosophy of science tended to be predicated on the assumption that science was conducted like Newtonian physics. Mayr, naturally, was arguing for an independent philosophy of biology—one that allowed for the historical contingency of evolution. Chemistry too has long been denied its own philosophy. If people thought about the issue at all, they typically assumed such a philosophy would somehow fit, a little messily, into the shape that physics had molded, with its reductive drive toward basic principles, its Popperian falsification of hypotheses, and so on. According to Scerri and Fisher the inception of a modern philosophy of chemistry only really happened in the 1990s, with the formation of the International Society for the Philosophy of Chemistry in 1997 one of the results.

My impression is that many practicing chemists are only vaguely aware their field has a philosophical aspect, and some might wonder why it needs one. Occasionally, though, chemical questions surface that reveal some underlying uncertainty in the discipline’s foundations and that can’t be resolved simply by appeal to objective laws and principles. One such question has been broached by Scerri, who has long been concerned with the fine details of the periodic table. It is, he says, undeniably one of chemistry’s “big ideas,” and yet what shape it should have isn’t entirely clear. Many alternatives to the conventional rectangular-block shape have been proposed, whether the “long form” with the lanthanide and actinide series inserted or shorter forms in which those series compose a kind of free-floating footnote. But behind all the suggestions for elaborate spirals and three-dimensional representations are lingering questions about how to arrange the periodicity in the first place. Is hydrogen’s proper place really atop the alkali metals? And when the lanthanides begin below yttrium in the third column, do we put lanthanum or lutetium in the 18-column “medium form”?

Since it is unclear that any representation of the periodic table can be definitively proved the “right” one (though there may conceivably be a “best”), this issue seems indeed to be philosophical. So too is the nature of chemical bonding, which, perhaps even more than the periodicity of the elements, is the real bread and butter of chemistry. When atoms come together to form molecules, their electrons become distributed more or less throughout the molecule. How to carve up these “electron clouds” into particular bonds or orbitals is then somewhat arbitrary, but the matter has given rise to some of the hottest disputes in chemistry. The rivalry between the molecular orbital and valence bond representations of bonding in molecules, championed respectively by Friedrich Hund and Robert Mulliken and by Linus Pauling (who was never averse to controversy), is well known to chemistry students. They’re generally told the choice is a matter of taste and convenience, but that irenic resolution hasn’t stopped further arguments about the best way to think about chemical bonding. The late Richard Bader was combatively insistent that only his quantum theory of atoms in molecules (QTAIM), which defines a bond in terms of the shape and gradient of electron density, was the correct way to approach the issue. But this theory would classify as a “bond” some interactions between atoms that others would hesitate to designate that way.

What’s more, QTAIM turns a molecule into an assembly of nonoverlapping “molecular atoms” rather like the division of a continent into distinct nations, each with a unique character. Thus theoretical chemist Paul Popelier suggests there are millions of different species of carbon atom, each tailored by its particular molecular environment. If so, we can’t really speak of carbon atoms as having any specific character any more than we can speak about the personality of some generic person. This approach seems to challenge the kind of intuition about chemical properties of elements that most chemists rely on daily.

rev1_2.jpg Fall 2017

William Blake’s etching of Isaac Newton, 1795–1805.

William Blake’s etching of Isaac Newton, 1795–1805. Newton helped make physics the dominant science, and its philosophy, too, became dominant. But chemistry, says Ball, must have its own philosophy, one that takes into account the fact it is a craft as well as a science.

Wikimedia Commons/William Blake Archive/Tate Collection

As QTAIM implies, there’s not even universal agreement about what we mean when we speak of atoms and molecules. Can, for example, a single molecule be constituted from two rings that are merely mechanically interlocked, without any conventional chemical bonds between them? Or is that what some call a supermolecule? This lack of precise definition of terms isn’t unique to chemistry; it’s not uncommon to find that the more apparently fundamental the concept (energy and force in physics, gene and species in biology), the hazier the definition becomes. But that’s one reason we have philosophy: to wrestle with irresolvable problems.

Another general motivation for philosophical discussion in chemistry is the nature of causation. Science is predicated on ideas of cause and effect: imagining any kind of satisfactory explanation of a natural phenomenon is hard without them. The randomness and intrinsic probability that entered quantum mechanics—it’s not possible to say why any one of the possible outcomes of a measurement was observed in a specific experiment—seemed to challenge causality at the root of physics, which was precisely what disturbed Einstein so much. Biology, meanwhile, seems slowly to be coming to terms with the inadequacies of a reductionist gene’s-eye view of all causality in the living world. And chemistry has its own wrangles with causality. One can offer a thermodynamic explanation of why a reaction happened as it did: this reaction pathway led to the state of lowest energy (strictly, of lowest so-called free energy). But sometimes what happens is more a matter of kinetics: the outcome depends on what happens fastest.

And as philosopher and psychologist Rom Harré explains, a causal explanation might be given at many other levels instead, depending on what offers the most insight. Schoolchildren are told that carbon “wants” to form four chemical bonds and that electrons “like” to be paired up in orbitals. The result of a reaction, particularly in organic chemistry, might be explained according to a detailed mechanism of synchronized rearrangements of electrons. Ions—electrically charged atoms—stick together because “opposites attract.” Increasingly, especially in molecular biology, scientists recognize that a chemical process can’t simply be explained by considering the main protagonists (a drug binding to an enzyme, say) but must also take into account what the surrounding environment is doing, such as rearrangements and movements of water molecules. A causal explanation or mechanism is then usually a simplification of an incredibly complicated process, and some simplifications are more appropriate than others. Thus explanation has a philosophical aspect, in chemistry as elsewhere.

Further, every field of science develops concepts that have explanatory value specific to that discipline. As Nobel laureate Roald Hoffmann has pointed out, chemistry is particularly replete with fuzzy concepts that can’t be precisely defined in reductionist terms but are nevertheless indispensable to the craft: ideas like electronegativity (how much atoms “like” electrons), oxidation state (how many electrons an atom has lost or acquired), and valence (how many bonds it forms). Do these concepts have anything to do with the real world, or are they just heuristic tools? This is an old philosophical question about emergent properties and ways of conceptualizing the world. From an electron’s point of view the concept of a person is meaningless. Does that mean the concept really is meaningless?

Questions like this make it plain there is plenty to discuss in a philosophy of chemistry that doesn’t pertain to the philosophy of, say, physics, biology, or cosmology. This book offers a nice (if often rather high-level) introduction to many of those issues, but it is very much a partial survey reflecting the preoccupations of the editors. Eric Scerri’s journal Foundations of Chemistry, while by no means neglecting historical studies, tends to focus on chemical theory. And Grant Fisher’s interests lie with the abstractions of chemistry: theories, models, representations. But Hyle, the other major journal in the philosophy of chemistry, edited by Joachim Schummer, has a somewhat different flavor, which to my mind is more rooted in the doing than the being. For chemistry is a synthetic science: it is concerned with making. “The essence of chemistry,” says Nobel laureate Jean-Marie Lehn, “is not only to discover but to invent and, above all, to create.”

The neglect of that aspect in this book is reflected in the contribution by the late Estonian philosopher of science Rein Vihalemm, who argued that science can be separated into the physics-like exact, quantified, deductive disciplines and the natural-historical, descriptive, classificatory disciplines like geology and biology. Chemistry, he says, shares in both natures. Yes indeed—but until you admit a third category that is not about description at all but about creation, you are missing a huge part, perhaps the dominant part, of the subject.

This synthetic character is also the primary reason why the notion that quantum physics “explained” (and thus completed) chemistry is absurd. Because chemistry is an art and a craft as well as a “science” (or rather, because those things are a part of science, if only we could let go of the traditional physics-eye view), it has philosophical dimensions appropriate to the creative sphere. It has an aesthetic; it involves ethics. Chemistry is shaped not so much by what the world tells us as by human choices, and those choices inevitably embody particular philosophies. Why do we make this and not that? (Perhaps because we consider this more beautiful than that?)

Conventionally, this creative aspect is labeled applied science, which, as Peter Medawar pointed out long ago, sits lower on an implied hierarchy than the “pure science” of discovery. It is regarded as a kind of technology, hardly deserving of a philosophy. To the humanities, long familiar with the idea that art, literature, music, theater, and other human endeavors of creation are deeply imbued with philosophical themes and questions, this neglect must seem bizarre. But the answer to a “physics anxiety” in the philosophy of chemistry is not to seek for things physics still can’t “explain.” It is to recognize the objectives are not the same.