November 26, 2018, by Brigitte Nerlich

Blueprint, a broken metaphor?

Three things came together that made me write this post: observing an increased discussion of the blueprint metaphor in genetics and genomics around the publication of a book called Blueprint, reading an old article by George Gamow, and reading a footnote in a forthcoming book by Philip Ball entitled How to Build a Human.

The blueprint metaphor

The blueprint metaphor has been used in genetics and genomics for a long time. It emerged from mapping aspects of ‘a photographic print composed of white lines on a blue background, used chiefly in copying plans, machine drawings’ onto more abstract phenomena like ‘stories’, ‘organisations’ and even ‘genes’. Blueprint can therefore now be used to mean ‘something which acts as a plan, model, or template’ (Oxford English Dictionary), as for example in the phrase ‘genetic blueprint’. There is a problem with this mapping though, as it mostly conjures up the unproblematic creation, indeed ‘construction’, of a phenotype from a genotype or a linear link between a gene and a trait or function.

The costs and benefits of this metaphor have been discussed since at least the 1990s, when some social scientists argued that it framed a vision of genes and genomes that was too deterministic, while others pointed out that ordinary people might understand it in a more nuanced way.

Over time, developments in genomic science have repeatedly undermined notions of determinism. This meant that the blueprint metaphor lost its lustre and science communicators were told to avoid it because it is outdated. As Claire Ainsworth said in 2015, “[a]sk me what a genome is, and I, like many science writers, might mutter about it being the genetic blueprint of a living creature. But then I’ll confess that ‘blueprint’ is a lousy metaphor since it implies that the genome is two-dimensional, prescriptive and unresponsive.”

However, determinism has reared its head again and with it the blueprint metaphor. This was in the shape of a book by the psychologist and geneticist Robert Plomin entitled Blueprint (October, 2018) with the provocative subtitle: How DNA makes us who we are. In the Prologue to the book he states no less provocatively: “This DNA fortune teller is the culmination of a century of genetic research investigating what makes us who we are.” So there we have it again: DNA determines who we are. Of course, this is not quite what he says in the book as many people have pointed out. But that doesn’t distract from the fact of the book’s title implies it.

Even before the publication of this book, the blueprint metaphor provoked some discussion on twitter. I shall reproduce this twitter critique here, as it summarises nicely what scientists think is wrong with this metaphor, before turning my attention to an older, but forgotten, version of the blueprint metaphor which might be less contentious than its ‘modern’ version.

I’ll argue that the blueprint metaphor has two incarnations. While one gives a much distorted vision of the power of DNA, the other might provide a better understanding of how genes actually ‘work’, i.e. do their job in the cell. One is grandiose and wrong, the other workaday and quite helpful.

Given that genomic science, unlike its popular image, has embraced complexity, flexibility and even randomness and chance, one can ask, as Jon Turney did in a chapter for a book I co-edited: “How can language and indeed metaphor start to reflect this more fragmented, complex and context-dependent view of genes, which focuses no longer on what genes are but rather asks what they do within a biological system that changes and develops over time?” (Turney, 2009). This is a big question, which needs more than a blog post.

In the following I first report on a critique of the blueprint metaphor used to frame what genes or genomes ARE; second, I report on how the metaphor was used to frame what genes DO inside cells.

Genes are not blueprints

In July this year Simon E. Fisher wrote a long ‘thread’ on twitter explaining what’s wrong with the blueprint metaphor. Simon is an expert on language and genetics. I shall reproduce his ‘thread’ as a text here, with some paragraph throws thrown in and leaving out images of blueprints etc. (but see my featured image).

The ‘title’ of the thread was: “Your genome is not a blueprint. A thread about misleading metaphors in science communication.”

“DNA is often referred to as a ‘blueprint for life’. A blueprint is an architect plan, technical drawing or engineering design. Like a blueprint, DNA contains information to guide construction, in this case of a living organism. Beyond that, the analogy rapidly breaks down. For blueprints, there’s direct 1-to-1 mapping between each element of a design/drawing & its counterpart in the final constructed product.

‘DNA as a blueprint’ implies that individual genes show 1-to-1 mapping with different parts of a body and/or its functions. Not so. Many genes encode proteins. Their linear DNA sequence carries info for stringing together amino-acids (out of 20 different ones) in a particular order. This order determines how the string folds into a 3D shape. The shape determines the functions the protein carries out.

Your genome contains 20,000 protein-coding genes, with diverse roles; enzymes, hormones, receptors, structural proteins etc. They work together in complex networks, building & maintaining a living body of a myriad distinct cell types, with different genes switched on/off.

Molecular networks include genes that build proteins which regulate activity of other genes. Even hub genes don’t show 1-to-1 mapping to tissue/organism outcomes e.g. the PAX6 gene doesn’t direct eye development by itself & it also plays multiple other roles in the body.

Unlike a blueprint, we can’t reverse engineer a genome from the appearance of an organism or the organisation of its tissues, however exquisitely we examine it. (Cells carry their own copies of the genome & so we can directly access the DNA, but that’s a different issue.) Why isn’t reverse engineering feasible? Because genomes guide the building of bodies through networks of gene activity interacting with other intrinsic & external factors via developmental cascades.” (Here reference is made to the new book by Kevin Mitchell: INNATE].

Overall then, Simon comes to the conclusion that “the blueprint metaphor has almost no explanatory value & is eerily reminiscent of the homunculus fallacy in accounts of vision. (There’s no little person in your brain looking at patterns of light on your retina; & that would lead to an infinite regress.).”

As Massimo Pigliucci said in his blog Footnotes to Plato (in a post from 2017): “the blueprint metaphor is untenable and in fact positively misleading, and should be replaced by the concept of developmental encoding”. Whatever that may mean…

So what to do? In the last part of his thread, Simon Fisher suggests something that rather is difficult to do:

“Maybe, here we should move away from metaphors & just try to tell it as it is. The reach of genetics is extending into many walks of life with potential impact on disease, health, society & education. Now, more than ever, it’s crucial to communicate principles of gene coding to broad audiences in an accessible way, while avoiding broken metaphors.”

But is that metaphor really ‘broken’, especially when we think about protein-coding genes which, according to Simon, “work together in complex networks, building & maintaining a living body of a myriad distinct cell types, with different genes switched on/off”.

Blueprint as an overarching big metaphor should be declared dead. But there are uses of the blueprint metaphor at a smaller scale that might still be useful, especially when it’s used to explain what goes on in a human cell, that is, what genes do, rather than what genes are.

Genes are blueprints

Recently, I was trying to find some information about the influence of cybernetics on genomics, as that influence might pick up more dynamic aspects of what genes/genomes do, rather than what they are. Re-reading Matthew Cobb’s book Life’s Greatest Secret, I stumbled across this paragraph “[a]lthough Gamow did attend a 1956 conference on information theory in biology, there were no direct interactions of any significance with scientists working in either cybernetics or information theory, with the exception of von Neumann.” (p. 119). I became interested in Gamow.

Wikipedia says: “In 1953, Francis CrickJames WatsonMaurice Wilkins and Rosalind Franklin discovered the double helix structure of the DNA macromolecule. Gamow attempted to solve the problem of how the ordering of four different bases (adeninecytosinethymine and guanine) in DNA chains might control the synthesis of proteins from their constituent amino acids”.

In 1955 George Gamow published an article in Scientific American entitled “Information transfer in the living cell” and said: “The nucleus of a living cell is a storehouse of information.” (p. xx)

What is more, he linked information theory to emerging work on cellular automata and said that a cell was “a self-activating transmitter which passes on very precise messages that direct the construction of identical new cells.” (p. xx) How does this construction work?

It’s here that ‘blueprints’ come in. It should be stressed however that at the time the mechanisms of protein synthesis (coding) were still quite mysterious. So metaphors were all there was, plus advances in computing, cybernetics and information theory, of course. Gamow wrote:

“Comparing a living cell with a factory, we can consider its nucleus as the manager’s office and the chromosomes as the filing cabinets where all the production plans and blueprints are storied. The main body of the cell, its cytoplasm, corresponds to the factory area where workers are manufacturing the specified product from incoming raw materials. The workers in a living cell are known as enzymes. They extract energy from the incoming food, break down the food molecules and assemble the separated units into various complex compounds needed for the growth and well-being of the organism.” (Gamow, 1955: 70, emphasis mine)

Here blueprints are not blueprints with a big B, so to speak, they don’t specify a whole organism or a human’s ‘destiny’! These are blueprints with a small b, part of the working ‘machinery’ of a cell. Their work is intricately linked to the information they convey. That in turn leads to questions about the ‘language of genes’, a language that was only just being understood by people like Crick, Watson, Franklin and many others.

“The information in the chromosome blueprints must be communicated to the working enzymes. How is the DNA language of the chromosomes translated into the protein language of the enzymes? What mathematical trick makes it possible to transform a message represented by a sequence using four symbols into one using 20 symbols?” (pp. 72-73, emphasis mine)

As Bradon Smith said in an article on metaphors in genetics: “It is interesting to notice the way in which in this passage the word ‘translation’ acts as a metaphoric hinge: the metaphor of a digital, number system shifts, via the concept of translation, to a metaphor of language – a protein ‘word’ formed of amino acid ‘letters’. The code metaphor, then, morphs into the language metaphor, especially in the immediate aftermath of Watson and Crick’s discovery of DNA in 1953.” There is another blog post in there!

On the whole Gamow’s article based on the code and language metaphor of ‘life’. But it also exploits a metaphor based on something Gamow loved, namely: card games. Lily Kay in Who Wrote the Book of Life calls it the cardgame of life metaphor. The blueprint metaphor links up with all of these other grander metaphors as well as with this smaller one.

“Suppose that each amino acid possesses a special chemical affinity for a certain triplet of nucleotides; that is, a given letter always tends to associate itself with a certain hand – a with three hearts, b with three diamonds and so on. Then the order of the nucleotides in a DNA molecule will uniquely determine the order of the amino acids in the protein built according to its blueprint.” (p. 74, emphasis mine)

My final quote from Gamow’s article is the most interesting perhaps, as he uses it as a verb, thereby stressing the dynamic nature of the blueprint metaphor with a small b, which contrast with the rather static one of the blueprint metaphor with a big B.

“Similarly in a living organism over eons of time the random mutations may once in a great while produce a sequence of nucleotides which blueprints a new and helpful enzyme. A lizard living in the age of reptiles may thus have acquired an enzyme which catalysed the production of mile – and so taken the first step toward the age of mammals.” (p. 78, emphasis mine)

Responsible Language Use

In its most popular use, the metaphor of the blueprint is misleading, as it jumps from genes straight to the whole organism, hiding all the dynamic work that goes into this construction. There are, as Simon Fisher has said, multiple interacting networks at work, there is development over time, there is dynamic and context-dependent folding and unfolding etc.

All this time cells are hard at work behind the scenes ‘blueprinting’. In this context the much less popular blueprint metaphor is perhaps still useful as it helps us understand what genes do, rather than what they are in a grand but delusional vision of life. Keeping this in mind, it might be a good idea to tear up only one blueprint while making careful use of the other.

Overall then, we should not use the blueprint metaphor just because it is there, as that only entrenches a wrong view of what genes and genomes are and what they do, something that should be avoided at a time when more and more people have to think about their genes and genomes. This is a question of what I have called ‘responsible language use’, something just as important as responsible innovation! Whether telling things as they are is an option, is another question.

Epilogue – gene editing babies

I had just finished drafting this blog post when it was announced that a Chinese team of researchers had gene edited babies. If this story turns out to be true, it will be huge. One announcement said: “A Chinese researcher claims that he helped make the world’s first genetically edited babies — twin girls born this month whose DNA he said he altered with a powerful new tool capable of rewriting the very blueprint of life.” (emphasis mine)


Image: Architectural drawing, 1902, Adrian Michael Wikipedia






Posted in genomicssynthetic biology