Seeking Answers: An Interview with Roald Hoffman

by Lesley A. Iwanejko

LI: Do you think scientists and poets are seeking answers to the same questions? Perhaps they are just entering and exiting at different spokes of the same wheel? Similarly, do you think science complements or influences poetry, or vice versa?

RH: To some extent we seek the same answers, and I would add to science and poetry religion too. We want to understand the beautiful and terrible universe around us and within us. True, the testing ground of the poem is the contact it makes with the emotions, and that of science more in touching reality through experiment. But there are many similarities, and some differences.

For me science influences poetry, through suggesting metaphor, and in the interesting language of science. I think that the conciseness of poetry has crept into some of my best science -- or maybe it was there. I would say it's a love for language, more than poetry, that has come into the science.

LI: I am interested in exploring the differences, if any exist, between people who primarily are 'professional' poets and scientists who also write poetry. Do you think that poetry written by scientists is 'different' to that written by other poets?

RH: If the poetry is good, it is no different. There are sources of metaphor and language, and some barriers (of having the cognitive background to recognize the poetic turn within a scientific field) it is an interesting question to ponder if there is science, theory and experiment, that is in some way poetic within the science. I have to think about that.

Is Chemistry An Art?

LI: I was also talking with my colleague Andrew Carnell in the chemistry department. We were looking at what seemed to be different ideas concerning how to 'do' science between biology and chemistry. We have a number of collaborative projects and it seems to me that the main aim and interest for an organic chemist is in finding a way to 'create' something, whereas, for me it is trying to understand something. Where our two areas do converge is wondering why what we predicted to happen does not actually happen. For me, I would then have to try to find out why but it seems that often in organic chemistry that is of lesser importance than adding the newly discovered route to the database and then recommencing the search for the synthetic route to the desired product. Is then organic chemistry, synthetic organic chemistry, more of an art than a science?

RH: It is both. But certainly the dominance of the act of creation makes organic chemistry different. And you've probably encountered people who keep asking "why" in order to avoid ever doing anything, who, I would say are afraid of the existential act of creation. This is part of my struggle with reductionism in the hands of physicists. Take a look at the chapter on that near the beginning of my book The Same and Not the Same, and the couple of chapters on Creation and Discovery later on.

LI: When you are thinking about and doing science, are you thinking in a different way to when you are thinking about and writing poetry?

RH: Hard to say. Some patterns of structure and organization are shared, and maybe that is a reason why I am not so good a poet as a scientist, I try to carry things over. Both are done in riffs and pieces of time, fragmented. I do have to separate myself from the day to write poetry, though that was not always so. In my science and the poetry I am always thinking as much of speaking to others as I am of explaining to myself.

LI: To build on that question, your experiences in Poland during the war have obviously and not surprisingly had a very strong influence on much of your poetry do you think you can separate the poetry that is about science from that which reflects your other life experiences, or is it impossible to separate the different aspects and how they impact on your poetry? Lesley: To build on that question, your experiences in Poland during the war have obviously and not surprisingly had a very strong influence on much of your poetry do you think you can separate the poetry that is about science from that which reflects your other life experiences, or is it impossible to separate the different aspects and how they impact on your poetry?

RH: I try not to separate my worlds too much; so the poetry touches on all the things in my life. Perhaps that is why writing in general and poetry in particular are so nice; they allow me to compartmentalize less. In the science I can't talk about those wartime experiences.

LI: Does writing about your feelings, experiences during the war help you to come to terms with what happened? In that sense, does poetry offer something more than just prose, a factual account if so are you able to describe what it offers?

RH: Sure, it helps to come to terms. But two things I feel strongly about: 1. Writing poems does not make the poet a better, more ethical person. That is a romantic fallacy. You can hurt people just as much with a poem as with a molecule. 2. There is difference between poetry and therapy. Poetry may be therapeutic for the writer, to be sure. But it is written out of yourself, for others (reflect -- you like poems; would you get to read them if they were not disseminated by their writers in some way?). Poetry, and science, are, as Derrida said of writing "the message that abandons".

Roald Hoffmann also answered the following question, posed by another member of staff at Liverpool, which we have also included:

About Thinking

Professor Steve Holloway (

Asks about: “[Roald] Hoffman’s belief in deploying simplistic theoretical methods to explain concepts.”

RH: Some would say too simplistic. Here is what a congenial reviewer, a friend, said on one of my papers:

"Somehow the problems facing an author writing for Accounts of Chemical Research are similar to those we meet in the sexual education of our children. To begin with you have to give them the basic facts of life in a grossly simplified way, skipping the finer technical details and the complications that usually contribute to the interaction . After that there are two possibilities: (a) let the little dears find out for themselves, or, (b) tell them a bit more about these problems. From the "Statement of Policies and Procedures" I see that the editors would prefer course b. However your contribution falls clearly into category a.... 

And concerning: “His book on surface electronic structure captures the flavour of this perfectly. [Steve] would like to know how [Roald] thinks the widespread acceptance of density functional theory will change our teaching of bonding in molecules (of all sizes) and solids and surfaces. We are now in a position where we have a computational method (DFT) which in many instances can quantitatively capture complex bonding arrangements. Should we be making more use of this at the undergraduate level to explore covalent and ionic bonding (for example)? Should we spend less time on wavefunctions and more time on electron densities as vehicles for conceptually developing the subject? What do you think?

Roald: My strong feeling is that most of these calculations provide an approach to simulation, with little understanding. Moreover, the man-machine interactions inherent in complex calculations make it psychologically unlikely that most of the people who do the calculations will seek understanding, or believe it can be attained. To some extent this is an epistemological question -- what is understanding? For the moment we see two responses in the literature: the tacking on of simpler MO explanations (of the type I do) onto the DFT calculations. Another strategy is just to abandon all attempts at interpretation, just giving numbers.

No, I don't think we should spend more time on densities. They are dull, looking just like a spacefilling model; and building up molecular complexity with them is the equivalent of building a ball and stick model. The application of a so-called Morokuma partitioning (sorry for the technical stuff here...) to DFT results is a good direction to go. ELF and Bader analysis in my opinion are of little value -- they are analytical and descriptive, and interesting in that way. But these methods lack the predictive and productive (suggesting experiments) aspects of a good theory, which is what qualitative molecular orbital theory provides.

I've written a bit more on the epistemological questions in Qualitative Thinking in the Age of Modern Computational Chemistry, or What Lionel Salem Knows, R. Hoffmann, J. Mol. Str. (Theochem),424, 1-6 (1998).

An essay by Roald Hoffmann, "Science, Language, Poetry", can be found at:

“Roald Hoffmann’s land between chemistry, poetry and philosophy” can be found at:

For further information on the work of Roald Hoffmann see:

Biographical Details

Lesley Iwanejko lectures in Genetics at the University of Liverpool. Her interview with Nobel Prize winning chemist Roald Hoffmann took place at the end of January 2006. Here Roald Hoffmann talks about the connections he makes between his science and his poetry. We are also delighted to be able to publish two of Roald Hoffmann’s new poems, one of which has appeared previously on this site.