Metaphor, alchemy and lessons from the 17th century

Philip Ball has just published a magnificent book on the history of alchemy: Alchemy: An Illustrated History of Elixirs, Experiments, and the Birth of Modern Science. This made me think about metaphor, of course, given how central metaphorical language was to alchemical practice.

In a sense, metaphor is alchemy, metaphorically speaking, as it transmutes two concepts into one. Magic! But that’s not what I’ll be talking about.

Reading a review of Ball’s book calling alchemy “the ultimate fool’s errand”, I remembered that during the 17th-century rise of empiricism in philosophy, both alchemy and metaphor came under attack, sometimes together, sometimes alone, and I wondered whether we can still learn something from these old debates about science and language.

There has been a strand of thought in the history of science and metaphor according to which “During the rise of empiricist epistemologies in the sixteenth and seventeenth centuries, metaphors suffered one beating after another at the hands of ‘scientific-minded’ philosophers” (Johnson, 1981: 11); and that modern empiricism subscribed to a “literalist dogma”, according to which “all informative language is univocal, that is, literal” (Hesse, 1987). However, there is another strand of thought according to which this is too simplistic a view. Scholars like Nicolaas T.O. Mouton and Brian Vickers before him, claim that classical empiricists did not reject metaphor outright as pure nonsense. Rather, they criticised what they saw as the excessive or unreasonable use of metaphor, particularly by some alchemists and occultists, while acknowledging metaphorical language is indispensable in many domains of inquiry, including science and philosophy.

In this post I’ll explore first what 17th-century natural philosophers grappling with the emergence of empiricism had to say about metaphor and alchemy and I’ll then try to draw some conclusions for the the present, when we are still grappling with the status of metaphor (and alchemy?) in science.

Metaphors and natural philosophy

In the following I’ll draw on insights by Nico Mouton and others on the relations between alchemy and language in the 17th century. Nico contributed a chapter to a book I edited with Armin Burkhardt back in 2010, entitled Tropical Truth(s): The Epistemology of Metaphor and other Tropes and I shall draw in this quite heavily for inspiration.

The use and abuse of metaphors

We all know, sort of, that John Locke, one of the foremost empiricist philosophers of the 17th century, described metaphors as ‘perfect cheats’ in his 1690 work, An Essay Concerning Human Understanding (3.10.34). You might also have heard about Thomas Sprat, historian of the Royal Society (founded in 1660), calling metaphors a “trick”, a ‘beautiful deceit” and even an “open defiance against reason” (1667, 1.25). That seems pretty damning. And it gets worse.

Listen to Samuel Parker, Bishop of Oxford, writing in 1666: “All those Theories in Philosophy which are expressed only in metaphorical Termes, are not real Truths, but the meer products of Imagination, dress’d up (like Childrens babies) in a few spangled empty words …. Thus their wanton and luxuriant fancies climbing up into the Bed of Reason, do not only defile it by unchaste and illegitimate Embraces, but instead of real conceptions and notices of Things, impregnate the mind with nothing but Ayerie and Subventaneous Phantasmes.” Talk about using ‘metaphorical terms’ and ‘ayrie and subventaneous phantasms”!!

But things are more complicated than they seem at first glance. Empiricists and natural philosophers did not just reject metaphors, they had quite nuanced views about their pros and cons. Locke acknowledged that it is scarcely possible to speak of the mind and immaterial matters without recourse to metaphor and using words like ‘apprehend’, ‘adhere’, ‘conceive’, ‘instil’, and so on etc., as our language for such abstractions emerges from comparisons with the sensible world (see Essay, 3.1.5). In 1674 the Jesuit philologist Pierre Besnier noted that “if we compare them to their first origin, most of our words are nothing but metaphors” (quoted in Aarsleff, 1982: 82, footnote 74).

Sprat, in turn, said about the body: “For, though Man’s Soul, and Body are not onely [sic]one natural Engine (as some have thought) of whose motions of all sorts, there may be as certain an accompt given, as those of a Watch or a Clock” (1.16) using contemporary mechanical metaphors and analogies. He also noted that tropes can “represent Truth, cloth’d with Bodies; and […] bring Knowledg [sic] back again to our very senses, from whence it was at first deriv’d to our understandings”. (1667: 1.25)

On the whole though, 17th-century philosophers preferred (pace Parker!) what Sprat called “a close, naked, natural way of speaking; positive expressions; clear senses; a native easiness: bringing all things as near the Mathematical plainness, as they can: and preferring the language of Artizans, Countrymen, and Merchants, before that, of Wits, or Scholars.” One can sympathise with that!

The dangers of metaphorical excess

By contrast, alchemical and occult traditions exemplified what empiricists saw as metaphorical excess, extravagance and even chaos, using a proliferation of terms and analogies without careful distinction, often leading to cognitive and epistemological confusion. What is more, they sometimes obfuscated things intentionally by the use of ‘cover names’ or so-called ‘Decknamen’, to keep outsiders from gaining easy access to their knowledge and preserve trade secrets (for example, the ‘red lion’ can represent ‘sulphur’, but things are much more complicated!). Lawrence Principe has discussed Decknamen at length and “how the unique lexicon and iconography of alchemy serve as both a gateway and a barrier to understanding its secrets”.  

Because of all this, Locke liked to distinguish between sound judgment (which starts with analysis and separation of ideas) and wit (which indiscriminately combines ideas via similarities); abuses in alchemical metaphor fell into the latter category. He condemned the use of “affected obscurity” (3.10.3) and new words without clear ideas attached, which were hallmarks of alchemical texts.

In his famous book The Sceptical Chymist (1661) Robert Boyle, natural philosopher, chemist, physicist, alchemist and “prolific analogizer” (Gentner & Jeziorski, 1993:457), accused other alchemists of playing too freely with names and metaphors (giving “divers things one name” or “one thing, many Names”), which results in confusion rather than insight. He also pointed out that “apposite comparisons not only give light, but strength to the passages they belong to, since they are not always bare pictures and resemblances, but a kind of argument”.

Metaphors between tool and trap

In contrast with the legitimate uses of metaphors to make sense of the world, 17th-century natural philosophers were worried about the chaotic and overextended use of metaphor in some alchemical and occult writings, where metaphors became so rampant and literalised that they blurred distinctions between sense and nonsense.

As early as 1619, Daniel Sennert, physician and, like Boyle and many others, an alchemist, commented on the alchemists’ (ab)use of the Microcosm-Macrocosm analogy and said: “the Analogie of the great and little World is extended too large by the Chymists, because they make not an Analogie, but an identity, or the same thing” (quoted in Haack, 1994: 4). In a way, collapsing a metaphor or similarity into literal identity forfeits its cognitive value and therefore constitutes an abuse of metaphor.

One can argue that for natural philosophers, metaphors exist on a spectrum from cognitive tool, when used with judgement, to cognitive trap, when used with mere wit; they can be used to obscure and obfuscate or to enlighten and illuminate; they can generate opacity or understanding; they can be productive or problematic. On the one hand they are necessary for dealing with the limits of direct experience but should not replace rigorous analysis or be used to excess. 

Empiricist philosophers did not criticise alchemists for using metaphors as such but for using them wrongly, that is, extending them beyond reasonable bounds, proliferating them or collapsing them into the literal. It should be stressed that although they thoroughly criticised the linguistic contortions in which some alchemists engage, philosophers like Locke and Boyle (and indeed Newton) were not averse to dabbling in alchemy and using its tools to advance knowledge about nature.

This is all very interesting you might say, or indeed very boring, but what can we learn from this for metaphor use in modern science? In the following, I’ll try to draw out some 17th-century lessons for metaphors use and abuse in modern genomics.

Metaphors and genomics

I have often talked about metaphors like the genome as a code or a blueprint and their pros and cons in scientific thinking and science communication. Are such metaphors obscuring or illuminating? Following in the footsteps of 17th-century philosophers, I’d argue that it all depends. They discovered that one needs metaphors on the one hand to make sense of a world beyond our senses, but that one can overindulge in them on the other and use them to produce utter nonsense – and everything in between.

Let’s look at some classical modern metaphors, such as the code and blueprint metaphors in genomics and see what 17th-century natural philosophers might have made of them.

The code metaphor: Insight or illusion?

When Francis Crick wrote a note to his son about the discovery of the structure of DNA in 1953 comparing it to a code, he and others probably thought this was an illuminating way of thinking about DNA, and, over time, this became the defining metaphor for genetic and genomic thinking.

This reminds me of Boyle defending the cognitive value of “apposite comparisons” and “analogous instances which do declare the nature or way of operating of the things they relate to” (quoted in Mulligan, 1994: 254) and Locke saying that “In things which sense cannot discover, analogy is the great rule of probability” (Essay, 4.16.12). Crick’s ‘code’ insight also demonstrates what Boyle meant by metaphors serving as “a kind of argument”, as it genuinely revealed structural relationships between DNA sequences and protein synthesis that might not have been as apparent through purely literal description.

This also maps quite neatly onto Locke’s distinction between wit and judgement. The ‘code’ metaphor works well when accompanied by careful analysis of where the analogy holds and where it breaks down (judgment). It becomes problematic when people “run immediately after” (Locke, 1697/1706/2017: 29) similarities without noting crucial differences – like the fact that genetic ‘codes’ are context-dependent, probabilistic, and subject to complex regulatory networks rather than simple one-to-one mapping like human codes.

In this way genomic metaphors can create an illusion of understanding molecular biology while obscuring the actual complexity of gene regulation and expression. Turning from science to science communication, there is a danger that such metaphors may give lay audiences enough conceptual purchase to feel they understand genetics, but not enough to grasp the actual complexity – creating a false sense of comprehension.

That danger is particularly acute when people make deterministic inferences based on the metaphor used in science communication which scientists and genetic counsellors, for example, then have to dispel. This makes me think of Locke when he warned in 1697 that similes are great for getting across an idea but we should be aware of the fact that similes may make “men think that they understand better because their hearers or readers understand them better” (1697/1706/2017: 28). 

The blueprint metaphor: Analogy or identity?

Such dangers are even more apparent when it comes to the famous blueprint metaphor of DNA, where, just like Boyle did with alchemical metaphors, Philip Ball worries that it may collapse distinctions between literal and figurative claims. I think, as with the code metaphor, one should consider intent, awareness and audience in the use of such a metaphor. In their daily work scientists use the code/encode, and sometimes even the more contentious blueprint metaphor as a tool for research and/or communication, while remaining, on the whole, aware of its limitations, whereas public discourse may treat such metaphors as literal truths on the one hand or a secret, arcane or even occult jargon on the other, like the deliberate use of coded language by alchemists.

Genomic metaphors: Plurality or proliferation?

When it comes to genomic metaphors more generally, such as the code, the programme, the blueprint, the book of life, the jigsaw, the language of God and so on, Locke and Boyle would be wary of this proliferation of metaphors and, in some cases, exaggeration. They might be reminded of the alchemist William Gratacolle’s (1651) endless synonyms for the Philosopher’s Stone which created an illusion of systematic knowledge while obscuring the underlying chemistry. 

Towards metaphorical literacy

I think modern critics of metaphors in science could learn from 17th-century empiricists who did not reject metaphors outright but argued that metaphors can be useful if their figurative nature is recognised and if we avoid the (alchemical) proliferation of metaphors and the confusion of similarity or analogy with identity.

Knowing a bit about the history of the philosophy of metaphor might contribute to what one might call ‘metaphorical literacy’, the ability to use metaphors as cognitive and heuristic tools while maintaining awareness of where they break down. The have a legitimate place in scientific discourse, as long as we are aware of their limitations.

This has perhaps some parallels with alchemy itself, where the valuable aspects of alchemical practice (experimental techniques, apparatus, chemical processes) should be separated from the problematic ones (mystical obscurantism, arcane language and the deliberate confusion of literal and metaphorical claims).

Epilogue

“I can’t resist ending this post with a quote from Brian Vickers (1984) on how 17th-century occult and experimental scientific traditions deal with language, especially metaphors and analogies. He claims that “a spectrum of beliefs and attitudes can be distinguished, a continuum from, say, absolutely magical to absolutely mechanistic poles, along which thinkers place themselves at various points depending on their attitudes to certain key topics. One of these topics, not much discussed so far, is the relationship between language and reality. In the scientific tradition, I hold, a clear distinction is made between words and things and between literal and metaphorical language. The occult tradition does not recognize this distinction: Words are treated as if they are equivalent to things and can be substituted for them. Manipulate the one and you manipulate the other. Analogies, instead of being, as they are in the scientific tradition, explanatory devices subordinate to argument and proof, or heuristic tools to make models that can be tested, corrected, and abandoned if necessary, are, instead, modes of conceiving relationships in the universe that reify, rigidify, and ultimately come to dominate thought. One no longer uses analogies: One is used by them. They become the only way in which one can think or experience the world.” (Italics mine)

Somehow that reminds me of today, where some no longer submit the language we use to argument and proof but let ourselves be used by language, not so much in the form of analogies, but lies.

Image: Wikimedia Commons Alchemist Sendivogius (1566–1636) by Jan Matejko, 1867 – I chose this because there is a woman in the picture!

Selected further readings

Primary sources 

Boyle, R. (1661). The Sceptical Chymist: or Chymico-Physical Doubts and Paradoxes, Touching the Experiments Whereby the Vulgar Spagirists are Wont to Endeavour to Evince Their Salt, Sulphur and Mercury to be the True Principles of Things. To which in this Edition are subjoyn’d divers Experiments and Notes about the Producibleness of Chymical Principles. London.

Locke, J. (1690). An Essay Concerning Humane Understanding. London.

Locke, J. (1697/1706/2017). The Conduct of the Understanding. Edited by Jonathan Bennett. In: Early Modern Texts in 2017. Written in 1697; first published posthumously in 1706. 

Gratacolle, W. (1651). Five Treatises of the Philosophers’ Stone. London.

Parker, S. (1666, 2^nd ed.). A Free and Impartial Censure of the Platonick Philosophie. Oxford.

Sennert, D. (1619). De chymicorum cum Aristotelicis et Galenicis consensu ac dissensu liber. (“On the agreement and disagreement of the chymists with the Aristotelians and Galenists”). Wittenberg.

Sprat, T. (1667). The History of the Royal Society of London for the Improving of Natural Science. (Sect. XX. Their manner of Discourse). London. 

Secondary sources 

Aarsleff, H. (1982). From Locke to Saussure: Essays in the study of language and intellectual history. Minneapolis: University of Minnesota Press. 

Ball, P. (2025). Alchemy: An Illustrated History of Elixirs, Experiments and the Birth of Modern Science. New Haven, CT: Yale University Press.

Ball, P. (2025). Was alchemy a load of nonsense? Substack: And Another Thing, 16 September.

Ball, P. (2006). Alchemical culture and poetry in early modern England. Interdisciplinary Science Reviews, 31(1), 77-92.

Clark, S.H. (1998). ‘The whole internal world his own’ : Locke and metaphor reconsidered. Journal of the History of Ideas, 59: 241-265.

Gentner, D., & Jeziorski, M. (1993). The shift from metaphor to analogy in Western science. In Ortony, A. (ed.), Metaphor and Thought. Cambridge Cambridge University Press, pp. 447-480.

Haack, S. (1994). “Dry Truth and Real Knowledge”: Epistemologies of Metaphor and Metaphors of Epistemology. In: Hintikka, J. (eds) Aspects of Metaphor. Synthese Library, vol 238. Springer, Dordrecht.

Hesse, M. (1987). Ayer and the philosophy of science. In B. Gower (Ed.), Logical Positivism in Perspective (pp. 69-88). Totowa, NJ: Croom Helm.

Johnson, M. (1981). Introduction: Metaphor in the Philosophical Tradition. In M. Johnson, (ed.), Philosophical Perspectives on Metaphor. Minneapolis: University of Minnesota Press, pp. 3-47.

Mouton, N. T. O. (2010). Metaphor, empiricism and truth: A fresh look at seventeenth-century theories of figurative. Burkhardt, A. and Nerlich, B. (eds.), Tropical Truth(s): The Epistemology of Metaphor and Other Tropes, pp. 23-49.

Mulligan, L. (1994). Robert Boyle, “Right Reason” and the meaning of metaphor. Journal of the History of Ideas, 55, 235-57.

Nerlich, B. & Clarke, D. (1996), Semantic Theories in Europe, 1830-1930. Amsterdam: John Benjamins.

Principe, L. M. (2012). The Secrets of Alchemy. Chicago: University of Chicago Press.

Skouen, T. (2011). Science versus Rhetoric? Sprat’s History of the Royal Society Reconsidered. Rhetorica: A Journal of the History of Rhetoric, 29(1), 23-52.

Vickers, B. (1984). Analogy versus Identity: The Rejection of Occult Symbolism, 1580-1680. In Vickers, B. (ed.), Occult and Scientific Mentalities in the Renaissance. Cambridge:: Cambridge University Press.

Vickers, B. (1988). On the function of analogy in the Occult. In Merkel, I., & Debus, A.G. (eds.), Hermeticism and the Renaissance: Intellectual History and the Occult in Early Modern Europe. Washington: The Folger Shakespeare Library.

Vogt, P. (1993). Seascape with Fog: Metaphor in Locke’s Essay. Journal of the History of Ideas. 54:1-18.

Wootton, D. (2025). Alchemy – the ultimate fool’s errand. The Spectator, September.