Entanglements between language,science,and worldviews

This issue of Interdisciplinary Science Reviews focuses on Evelyn Fox Keller's work on the various roles that metaphors play in science and, especially, in genetics and developmental biology (Keller 2015). Using her paper ‘Cognitive functions of metaphor in the natural sciences’ as a springboard, a collection of scholars – historians, philosophers, and sociologists of science, as well as theoretical biologists, a theoretical physicist, and an anthropologist – reflect on the role of language in constructing, sustaining, and challenging scientific ideas and worldviews.

It is highly fitting that Keller's work should be celebrated in Interdisciplinary Science Reviews. One constant theme running throughout a prolific and wide-ranging career is her commitment to interdisciplinarity: how insights gained and combined from multiple perspectives can enrich our understanding of science, its practice, its structure, and its place in society. Keller trained in theoretical physics, molecular biology, and the history and philosophy of science. All these influences exist mutualistically in her work.

Keller has made pivotal contributions to our understanding of the role of language and metaphors in science as well as the pervasive influence of gender in the way many scientists have approached and represented nature. Her 1983 book A Feeling for the Organism: The Life and Work of Barbara McClintock constituted one of the earliest and most influential contributions to the history of women in science and to feminist studies of women's accomplishments. Keller showed how Barbara McClintock's vision of science shaped her creative approach to the cytogenetics of maize and enabled her to discover the ‘transposition’ of genetic elements, a key step in attempts to understand the complexity of the genome. In her Reflections on Gender and Science (1986), Keller pointed out the limitations in the dominant conception of objectivity in Western science. 1 Within this framework, the scientific approach to nature is static, tries to eliminate the subjectivity of the scientist, and focuses on control. Keller called for adopting a new approach: dynamic objectivity. Rather than trying to eliminate the knowing subject, dynamic objectivity encourages a fluid and interactive relation between the cognitive subject and the object of study. She continued her examination of how gender and language have shaped visions of science, its theories, and its practices from Francis Bacon to the present, in a series of studies collected Secrets of Life/Secrets of Death: Essays on Language, Gender, and Science (1992).

Keller then turned to analysing how metaphors have influenced the study of biology in Refiguring Life: Metaphors of Twentieth Century Biology (1995). Scholars had recognized the important role of metaphors in science (Hesse 1988; Young 1971), but here Keller offered a detailed analysis of how some metaphors used in embryology and genetics directed scientific inquiry in specific directions. In the last two decades Keller has published other important studies in the history and philosophy of biology, including: The Century of the Gene (2000); Making Sense of Life: Explaining Biological Development with Models, Metaphors, and Machines (2002), and The Mirage of a Space Between Nature and Nurture (2010). Her work on how self-organization is understood in biology has also been quite influential (Keller 2008, 2009). A surge of exciting research in biology has brought renewed attention to many of the issues Keller raised in those publications.

Recent discoveries about genetics and development have led to a revaluation of the ways that biologists had become accustomed to thinking and talking about genes, as master molecules, as constituting the blueprints for an organism (Barnes and Dupré 2008; Blaxter 2010; Bonduriansky 2012; Braun and David 2011; Francis 2011), and the units of selection in evolution. Environmental and developmental factors have been shown to interact with genetics in complex ways. Highly influential studies about phenotypic plasticity and epigenetic inheritance (Gilbert and Epel 2009; Hallgrimsson and Hall 2011; Jablonka and Lamb 2005, 2007; Mattick 2010; Müller 2010; West-Eberhard 2003; Sultan 2015) have demonstrated the need to rethink fundamental issues in biology, specifically what Keller described as the genotype-environment ‘entanglements’ (Keller 2010). These developments have led scholars to call for a new understanding of the role of organisms and agency in biology (Kauffman 2006; Walsh 2015; Sultan 2015). In addition, these novel approaches to understanding the natural world encourage us to revisit our notions of causation, and point to the need to generate more apposite metaphors that better capture the complex interactions taking place in biological systems in their ecological contexts (Huneman and Walsh 2017). The vibrant area of gene-environment interaction compels us to consider emerging interpretive frameworks for studies of evolution and genomics.

Furthermore, as Keller and other scholars have demonstrated, the language we use in scientific inquiry and the metaphors we adopt not only have an impact on descriptions of biological processes, but also on how scholars in other disciplines (Gibbon and Novas 2007; Pálsson 2007) and society at large interpret their significance. Several studies have explored how the view that genes determine an organism's characteristics has shaped understandings of race (Nelson 2016; TallBear 2013; Wailoo, Nelson, and Lee 2012), the nature-nurture debate, child development (Vicedo 2013, 2017), medical conditions and disabilities (Vicedo 2021), and other social issues (Lewontin 2000; Taussig 2009).

Thus, at this moment, exciting work in the sciences and in the humanities reveals the need to re-think many of the categories we have been using to do science and to interpret its epistemological and social value. Inspired by these new directions of research in biology and exciting work in science studies, the contributions in this volume testify to the fruitfulness of such interdisciplinary approaches.

The essays collected in this volume all, in one way or another, explore the role of metaphor, its capacity to inspire, constrain, expand, and mislead scientific thought and practice. Scientific discourse is saturated in metaphor. Yet, the significance of metaphor in science has been largely overlooked, or misunderstood. The power of metaphor to frame, constrain and inspire science has taken on particular resonance for biology in recent years, as evolutionary, developmental, molecular biology finds itself at a pivotal juncture. It is of critical importance to understand how the metaphors that permeated biology throughout the twentieth century have brought us to this juncture, and how new ones might lead us forward. In the target article, Keller addresses these fundamental issues.

Reflecting on the ubiquity of metaphor in scientific discourse, her ‘Cognitive functions of metaphor in the natural sciences’ examines ‘its seeming necessity.’ Metaphors, she claims, are necessary in scientific research ‘for generating knowledge about a world not yet known.’ They point to similarities and differences between what we know and what we don't yet understand. In doing so, they actively shape the direction of our thinking and shape our view of the world. So, she asks: ‘How is our scientific picture of the world shaped by our choices of metaphor?’

In this paper, Keller reflects on the role of metaphor in the history of genetics. First, she analyses the case of classical genetics. As she did in her book Making Sense of Life, Keller argues that the notion of ‘gene’ combined two different metaphors: the notion of the gene as the ‘atom’ of life and the gene as ‘an agent capable of directing the formation of particular traits.’ The gene, as she puts it, was supposed to be ‘simultaneously immortal and agentic’ in order to guarantee intergenerational stability and to guide individual development. The difficulty of solving the second question (how did a gene ‘determine’ a phenotypic trait?) led to ‘the discourse of gene action,’ a narrative that has shaped the field of genetics to this day. Yet, despite the fact that it had a ‘clearly identifiable effect on the course of research’ and helped the success of the field, Keller argues that ‘little of this work shed light on the question of how genes actually contributed to, guided, or shaped biological development.’ Nevertheless, the discourse of gene action played a critical role in shaping genetic narratives even after the advent of molecular biology.

Examining this role, Keller unearths ‘a conservative rather than innovative role, stabilizing rather than destabilizing’ role for metaphors and other tropes in science. Despite the great impact of molecular biology, Keller maintains that this field ‘failed to disrupt (indeed, even contributed to) the essential structure of the discourse of gene action on which classical genetics had previously been built.’ However, new contributions from genomics and developmental biology together with the impact of various cultural transformations have disrupted several assumptions underlying the belief in the power of genes and the discourse of gene action.

For Keller, recent biological discoveries demand a change in our views about the genome and a reconceptualization of the way we study living systems. Such a reconceptualization requires abandoning the simplistic dichotomies between nature and culture, the environment and the organism. As she puts it:

We have long understood that organisms interact with their environments, that interactions between genetics and environment, between biology and culture, are crucial to making us what we are. What research in genomics seems to show is that, at every level, biology itself is constituted by those interactions – even at the level of genetics.

In the spirit of Keller's work, the papers in this special issue encourage us to move beyond dichotomies to emphasize interactions and the complexity of living systems.

The first two papers situate some of the central issues discussed here in their historical context. This look at the past not only helps us to understand how we got here, but it also reveals that history itself has been simplified in the efforts to maintain a given narrative in biology. To re-examine that history is thus essential in order to offer a better historical account and to develop richer scientific research programmes.

In her paper, ‘Biology's Mistress, A Brief History,’ Jessica Riskin explores the complicated history that biology has had with teleology. Biology, it turns out, cannot live happily with teleology, but can't live without it either. As she notes, when the French naturalist Jean-Baptiste Lamarck coined the term biologie in 1802, he thought living beings arose from an innate striving impulse intrinsic to living matter. For him, agency and organization were defining aspects of living beings. That made them different from machines. In this way, as Riskin puts it, Lamarck ignited ‘one of the longest-lived controversies in the history of science.’ French comparative anatomist Georges Cuvier started it, but as the papers in this volume testify, it continues to this day. Riskin examines how Louis Pasteur's 1859 refutation of spontaneous generation and August Weismann's experiments cutting the tails of mice during 1887–1889 carried out to disprove belief in the inheritance of acquired characteristics were used against Lamarckian ascriptions of agency. For Riskin, ‘the “Weismann barrier” separating somatic cells from germ cells was a barrier separating individual agency and evolutionary transformation.’ Over time, it became ‘an unshakeable scientific axiom.’ Riskin's close analysis of the political and scientific beliefs of the actors involved in erecting and defending this axiom all the way to the present day shows how different authors used a simplified history to preclude a ‘serious consideration of what form a naturalized agency might take.’ Riskin calls for a ‘restoration of history to evolutionary biology,’ in the hope that it would help move us beyond the ‘dynamic stalemate’ of the discussion about teleology in biology.

Gregory Radick also shows how history can be useful in current scientific debates. He reflects on why talk of ‘genes for’ has been so resilient in his paper ‘Making Sense of Mendelian Genes.’ He argues that the ‘ambi-valence’ in a concept that was admittedly simplistic but capacious enough to encompass different levels of complexity made ‘the Mendelian gene – and the determinism it underwrites – so long-lived.’ He traces the emergence of the conception of genes as responsible for a phenotypic character to the work of William Bateson in England in the first decade of the twentieth century. Radick's close reading of the American T.H. Morgan, Alfred Sturtevant, Calvin Bridges, and Hermann Muller's 1915 book The Mechanism of Mendelian Heredity reveals that they already had a complex view of genes (which here they still called factors). In this canonical text, Morgan's group emphasized that talk of a gene or factor ‘for’ a visible character (e.g. eye colour), or for a version of a character (e.g. red eye) was just convenient shorthand. They noted that the genetic factors are influenced by many developmental and environmental elements, including temperature or age differences. The shorthand view was necessary ‘precisely because everything is interacting with everything else, all the time, with the most complex consequences.’ As Radick shows, these researchers were aware that many other geneticists were using the factor/gene as ‘a simple character-maker rather than a complex difference-maker.’

Radick argues that the ‘gene for’ locution is ‘straightforwardly shorthand for genes-as-character-makers talk and only unstraightforwardly shorthand for genes-as-difference-makers talk.’ Therefore, in an education system that emphasizes talk of ‘genes for’ despite disclaimers about the complexity of genetics and development, the heuristic power of this locution continues to shape scientific and popular understandings of genetics. In order to overcome this, Radick calls for fundamental changes in our educational system, so that the complexity of living systems is highlighted from the start.

The history of genetics thus shows us that language is difficult to contain, and expedient choices may have heady consequences in framing how research is conducted and living systems are understood. Once we adopt certain words and metaphors, they carry us, not always in directions we may have anticipated or wanted to go (Morgan himself often complained about how the success of genetics relegated embryology to the sidelines). To change directions, we might need new metaphors.

Papers by Fulda, Soto and Sonnenschein, and Walsh address the issue of the genetic programme metaphor in biology. In each case, the authors acknowledge the power and productivity of the programme metaphor in the growth of genetics, but each also stresses its constraining influence on biological thought. One common theme running through these three contributions – also reflected in Riskin's paper – is the concept of agency. Each of these three papers, in its own way, advocates shifting our attention away from genetic programmes and toward the agency of organisms.

In ‘Biopsychism: Life between Computation and Cognition’ Fermín Fulda documents the progressive movement away from the computational metaphor that promoted gene centrism toward the more recent emergence of the ‘cognitive metaphor’. Increasingly bacteria, slime moulds, plants, are credited with making ‘wise choices’, with deliberate planning, possessing ‘foresight’, and intelligence. This trend arises as a response to the evident inadequacies of programme talk. While the cognitive metaphor correctly highlights the constraints of the computational metaphor, it invites a certain sceptical incredulity by exaggerating organisms’ capacities as agents. As Fulda sees it, the problem lies not in the idea that organisms are agents, but in our conception of agency. It evinces a ‘Cartesian dichotomy’; everything in the world is either a mere machine or a full-blown cognizer. Generally, living things lie between these mechanist-intellectualist extremes. Fulda advocates an Aristotelian, ‘ecological’ view of agency. According to this approach, an agent is a system that responds to its affordances in pursuit of its goals. All organisms are agents, Fulda argues, but as the ecological approach shows, agency comes in degrees. Organisms with large adaptive repertoires inhabit richer systems of affordances, and are capable of pursuing an expanded range of goals more supplely. The ecological approach to natural agency frees us from the competing constraints of two extremal metaphors – the machine and the mind – and offers a more nuanced understanding of the agency of organisms.

Soto and Sonnenschein, in ‘Information, Program, Signal: Dead Metaphors that Negate the Normative Agency of Organisms’, echo the contribution of Longo and Mossio in noting the greater reliance on metaphor in biology than in physics. The information, programme, and computational metaphors, so prominent in biology, mislead in part because organisms inhabit a messy material world, not a pristine mathematical one. Furthermore, the authors point out, the programme metaphor invites a software/hardware dualism (reminiscent of the Cartesian framework discussed by Fulda) that is incapable of encompassing the ‘radical materiality of life’ (quoting Giuseppe Longo). The information/programme metaphor, Soto and Sonnenschein argue, is a dead metaphor. As a corrective, they seek to reintroduce an organicist perspective that harkens back to Aristotle (again) and to Kant. The organicist perspective underscores the importance of organisms as natural agents. In the organicist view, organisms are organizationally closed, far-from-equilibrium, historically contingent systems, permanently constructing and individuating themselves, capable of enacting and responding to norms. Any theoretically adequate account of the organism must be able to encompass these constitutive features of living things. These authors envisage such a theory being built around three guiding principles: (i) the principle of biological inertia, according to which the default state for a living thing is proliferation (including variation and motility); (ii) the principle of variation, according to which biological processes at all levels are source of variation, plasticity, and novelty, and (iii) the principle of organization, in which the closure of constraints is the ground of persistence and stability.

Walsh, like Soto and Sonnenschein, stresses that the genetic programme metaphor is a dead metaphor. In ‘Action, Program, Metaphor’ Walsh argues that the genetic programme metaphor is the victim of ‘vicious abstractionism’ (what Soto and Sonneneschein call ‘reification’), in which an abstract putative resemblance between genes and computer programmes becomes hypostatized as the shared essence of a kind that encompasses both. But if genomes really are programmes, they must possess the properties that confer on programmes their special explanatory privilege in software/hardware/environment systems. Walsh argues that this entails that genes must issue commands. Walsh borrows from recent work on natural semantics to argue that contemporary genomics and epigenomics suggest that genes typically do not carry the informational profile characteristic of directives (commands). If it can be said at all that any orders are being issued, it is developmental systems directing genes. In this respect, the genetic programme metaphor has been positively misleading. Like Soto and Sonnenschein, Walsh points out that those who initially promoted the genetic programme metaphor were motivated by an animadversion to natural teleology. In likening the genome to a programme they sought to situate the locus of executive control or agency in genes, on the one hand, and in the mechanisms of natural selection on the other. The programme metaphor does both harm and good: it is in Mary Hesse's words ‘both constraining and flexible’. On one hand, it mislocates the source of biological agency. On the other, it correctly suggests that making sense of life – explaining biological structure, function, development and inheritance – requires us to recognize a locus of executive control, or agency. As Fulda, and Soto and Sonnenschein argue, agency is inherent in organisms, not in genes.

Herrington and Jablonka explore the metaphors used in the articulation and promotion of the Extended Evolutionary Synthesis (EES) in ‘Creating a “Gestalt Shift” in Evolutionary Science: Roles for Metaphor in the Conceptual Landscape of the Extended Evolutionary Synthesis (EES)’. Like Soto & Sonnenschein and Walsh, these authors stress that the guiding principles behind the growth of twentieth-century evolutionary biology (the modern synthesis, or MS) rely on a literalistic, overly doctrinaire interpretation of its guidance metaphors. The overreliance on genes and programmes has marginalized organisms from participation in evolution. The ‘EES challenges all the assumptions of the MS that exclude or marginalize the role of development in evolutionary change’ (Herrington and Jablonka, 5). Here again, these authors, following Keller, highlight the death – the loss of ‘metaphoricity’ – of the guiding metaphors of the MS. They introduce the rhetoricians concept of ‘ekphrasis’, ‘speech that paints a picture in the minds’ eye … ’ to highlight the productive, exploratory, expansive role of metaphor in scientific reasoning. The power of ekphrasis is seen particularly vividly in the ways that EES proponents seek to inspire and articulate a ‘gestalt shift’, new ways of thinking about evolution, new ways of seeing biological phenomena. The authors note a proliferation of new metaphors intended to re-orient thinking about the role of development in evolution. The ‘music analogy’ is appealed to repeatedly. The allusion to music draws attention away from the script or score, and focuses on the performance of living. Music can be transmitted, and can evolve, through the copying of scores, but also through new ways of performance, new modes of recording. Spatial metaphors are often invoked (witness ‘Extended’ in EES) to illustrate the presumptive relation between MS and EES systems of thought. In extending the tools of evolutionary thinking, the EES also extends the domain of legitimate questions. Some opponents of EES express frustration at the inchoate nature of these metaphors. The metaphors don't issue directly in operationalized, testable hypotheses. Nevertheless, Herrington & Jablonka argue, the crucial contribution of metaphor in effecting a radical change in our conception of evolution is indispensable.

In ‘Geocentrism vs Genocentrism: Theories without Metaphors, Metaphors without Theories’ Longo and Mossio contrast the largely metaphor-free development of theories in physics (their example is geocentrism), with their metaphor-laden counterparts in biology (e.g. genocentrism). Metaphors play different roles in physics and biology. The machine, information, programme metaphors, Longo & Mossio argue, have not evolved into sufficiently robust theories, unlike their counterparts in Galilean and Newtonian physics. In fact, the same metaphors (e.g. the machine) that generated successful theories in physics have failed to do so in biology. Longo & Mossio argue that the programme and information metaphors were initially introduced into biology not as metaphors, but as fully actualized theoretical concepts imported whole from other sciences (Information Theory and Computer Science). They have not been successfully assimilated into the core of biological theory; instead they have come to be used merely as metaphors. Whatever formal and theoretical enrichment they might have provided as genuine theoretical concepts, this has been lost. Longo & Mossio illustrate the deficiencies of genocentrism with the cases of randomness and determination. Genocentric biology fails to mark the crucial theoretical difference between randomness and noise. As such, it has to concede that all uncertainty falls beyond its purview. But randomness (in contradistinction to ‘noise’) should be seen as an integral part of biological processes, and as such should figure in biological explanation. Similar considerations apply to the idea that genetic programmes determine biological form. As remarked by Soto & Sonnenschein and Walsh, the notion of a programme was initially incorporated into biological thinking as a way to circumvent the commitment to purpose. But here again, Longo & Mossio, also echoing Riskin, stress the need to acknowledge the ‘intrinsic purposiveness’ of organisms. Genocentric mechanism is incapable of explaining the salient features of organismal form, function, and dynamics. These authors conclude on a rather sceptical note concerning the capacity of informational and programme metaphors to generate a workable theoretical biology. These metaphors fail to accommodate the distinctive particularities of living things.

Tom McLeish draws his gaze beyond the particulars of the programme or code script metaphor and asks how this metaphor draws its strength from, and reflects back upon, a more encompassing metaphor, that of reading the Book of Nature. In his contribution, ‘A Meta-Metaphor for Science: The True and the Fictional within the Book of Nature’, McLeish notes that in early Christianity and the high Medieval period, the metaphor of the world as a book, and of science as the project of deciphering it, goes hand in hand with reading another book, sacred scripture, and is, complementarily, an act of piety. As literacy expanded in the Reformation, so too did the injunction to read and interpret nature's book; it became a task and a duty for everyone. McLeish documents the way that the metaphor of the book of nature changes along with the changing role and reach of print. But the metaphor of nature as book is beset by problems (‘four pitfalls’). The reading is context dependent. Moreover, the notion of the book of nature written once, unchanged in perpetuity, seems out of keeping with a dynamic evolving nature. Consequently, the project of pursuing science became one of writing nature (as the journal emerged), rather than reading it. The process of writing nature has developed a method, ‘methodological naturalism’, that allegedly isolates the process from undue influences of the scientist's extra-theoretic opinions – personal, aesthetic, and theological. McLeish argues that the metaphor of the code suffers these same pitfalls. It too has a cultural history; it too is open to be written; its subject matter is in no way independent of our conception of, and aspirations for, its subject matter – ourselves.

The previous papers highlight the diverse roles metaphors play in different sciences, the shortcomings of specific metaphors in contemporary biological research, and the important contribution that history can make to biology, a science dealing with historical beings and processes. They also present new proposals to enrich our current views in biology and in the history of science. Reflecting on some other aspects that we need to pay closer attention to, the following contributors add to those proposals. They call for embracing the creative instabilities of metaphor (Doyle), for adopting a more ecumenical ontological approach by incorporating the study of interactions (Longino), and for raising our awareness of the political ‘stakes’ of metaphors in science (Helmreich).

In ‘Jumping Concepts and other Transpositions: The Keller Effect in/on Discourses of Living Systems,’ Richard Doyle investigates the creative instabilities of language by looking at Keller's early work with Lee Siegel on slime mould aggregation (Keller and Segal 1970). Doyle argues that ‘the holistic coherence of a living system or a discourse becomes illuminated through its instabilities – the Keller Effect.’ Doyle believes that Keller's main contribution to our understanding of the role of metaphor in science is her insistence on how the ‘vitality’ of metaphors comes from their instability. Looking at the fascinating research of John Bonner on slime moulds, he points out that ‘John Bonner exists precisely due to the creative instability and hence semantic exchange that attends even the keywords “slime mold”.’ Thus, rather than a problem, the instability is an asset that opens creative possibilities. Doyle is interested in the ‘feedback loops onto knowledge production’ that may be enabled by ‘self aware collective metaphorical selection.’ While we may not completely choose our metaphors, we can embrace instability and raise our awareness of ‘the sheer necessity of the unknown to garner our attention, where it may possibly be transformed into the “not yet known”.

Helen E. Longino has addressed the question of how to best access the ‘not yet known’ by emphasizing, like Doyle, the importance of the collective input of the scientific community. In ‘Interaction: A Case for Ontological Pluralism’ Longino makes the case for studying ‘interactions’ and giving them the same ontological status as other entities in science. She calls for an ‘ontological pluralism’ that moves biology and other sciences beyond their narrow focus on ‘individuals.’ Longino argues that ‘a focus on individuals in some areas of analysis unwarrantedly limits the scope of investigation in those areas.’

Longino focuses on two research contexts where giving interactions the same status as individuals increases the scope of investigation: social epistemology, and social behaviour. In epistemology, instead of focusing on the individual cognitive agent one should focus on the community because questions such as what counts as evidence, what instruments and tools are appropriate in a given area, and others, are settled through the critical interaction among members of the research community. In the study of social behaviour, one can only see the full range of questions and grasp the significance of many phenomena when one looks ‘at the social behaviors as interactions, rather than as expressions of individual dispositions.’ In addition to those two areas, Longino points out that Keller has shown the significance of interactions in her own work about geneticist Barbara McClintock and developmental biologist Christiane Nüsslein-Vollhardt. As Longino emphasizes, interactions are central to living systems. She contends that interactions should play as fundamental a role in analysis and investigation as individuals as their states and properties do. She thus calls for an ‘ecumenical ontology,’ one that recognizes a plurality of ontologies and that is not driven by reductionist or unificationist aims.

In the last paper of the collection, Stefan Helmreich draws our attention to a crucial issue: despite the power of metaphors, the effects of science on individuals and societies are not metaphorical. In ‘Not a Metaphor,’ Stefan Helmreich highlights the ‘political stakes of metaphor.’ He first emphasizes the significance of Keller's focus on how language ‘with its ambiguity and multiple meanings – inescapably conditions scientific theory all the way through.’ He focuses on how metaphors engage in a ‘rhetorical oscillation akin to zeugma’ which refers to the use of a word to modify other words in a sentence but applying to them in different ways. Sometimes, a phrase is used in a literal way and in a figurative way in the same sentence, thus ‘destabilizing what was thought to be literal in the first place.’ Helmreich uses zeugma to explore the affective and the political aspects of metaphor. Building upon Keller's analysis of the ‘zeugmatic’ tensions in talk of genes as programmes, Helmreich draws our attention to another key aspect: the articulation of gene talk with the material culture of laboratories in ways that he describes as ‘material-zeugmatic.’ Following Donna Haraway, he proposes that the power of metaphors is shaped not only by language, but also by ‘hardware.’ In the lab, talk of ‘genes’ is inextricably tied to the material role in a specific experimental system. Therefore, he argues: The ‘gene,’ then, does not belong only to the realm of metaphor if by metaphor we mean ‘a figure of speech,’ but is rather, in an expanded zeugmatic sense, a ‘figure of speech, lab, and machinic writing.’

This implies that metaphors have ‘many determinations,’ in different realms, such as ‘media and technological artifacts or in political economy.’ Helmreich argues that paying attention to how metaphors operate in this ‘material-zeugmatically’ way can help in the area of transbiology and in the decolonization of genomes, since the use of genome sequences in human ancestry projects have often supported ‘notions of race-as-heredity that have been embedded in projects of colonial governance and subordination.’ He urges us to examine not only the ‘cognitive functions of metaphor,’ but also the ‘political functions of metaphor.’

Taken together these essays bring into relief a range of recurrent themes in Evelyn Fox Keller's work. They draw our attention to the subtle and intricate ways in which language, thinking, and doing are inextricably intertwined. The metaphors we employ enrich, expand, inspire, trammel and mislead. Many of the papers here directly address the place of metaphor as metaphor in scientific discourse. The contribution of metaphor isn't merely semantic. Metaphor frames our thinking and destabilizes it. It opens us to the unknown. It is a source of ‘creative instability’. It serves a cognitive function to be sure, but also a political one. It infuses scientific discourse with ‘affect’. Metaphors don't simply have localized, domain specific influences on science. The growth of science has itself been pursued under the auspices of a metaphor: reading the book of nature.

A number of the authors collected here sound a cautionary note about the influence of metaphor on scientific discourse. While metaphor clearly facilitates scientific investigation, we must also be cognizant of the power of metaphor to beguile and mislead. ‘Expedient’ choices, and convenient heuristics, have a tendency to become entrenched and hinder new developments. This pitfall of metaphors is perhaps nowhere more pronounced than in biology. The ineluctable deployment of metaphor in biology influences not just the way we speak about living things, but also how we think of the living, and how we practice biology. Or perhaps it is merely brought vividly into relief by the state of current biology.

As more than a few of these contributions highlight, the metaphorical nature of much of our scientific discourse about the living has a tendency to elide into metaphysical commitment. When the metaphoricity of metaphor is forgotten, its capacity to inspire creative thought, to adapt, to address the unknown, is diminished. As Keller's target paper attests (and as contributors agree), the metaphor of a genetic programme offers a salutary lesson in this regard. A number of papers in this collection stress that the genetic programme metaphor has become reified, has lost its essential metaphoricity. But as metaphysical doctrine, rather than metaphorical device, it now serves to hold biology back. It is difficult to square the idea that genes or genomes really are programmes, with the dynamic new biology of eco-evo devo, the Expanded Evolutionary Synthesis, Autonomous Systems Theory. At the same time, it must be said that the programme/information metaphor has helped to propel biology to the point where it is no longer applicable. The programme metaphor contains the seeds of its own obsolescence.

One prominent theme running through this collection is the sheer complexity of living things. Life is very different from the non-living world. This has always raised a problem for any scientific enterprise that seeks to articulate the common themes in nature. Biological phenomena, taken in their full, dynamic, supple adaptiveness, are difficult to encompass within the bounds drawn by the sciences of non-living phenomena. Metaphors help us to make sense of life; they help us to adapt the tools we bring to bear in understanding the non-living to encompass the phenomena of the living. Several papers in this collection discuss the way that metaphors allowed biology to borrow concepts from other sciences, including computer science, information theory, and cybernetics. Other authors discuss the way in which the programme metaphor served as a device for circumventing the need to appeal to the very properties that mark out living things as distinctive: their purposiveness. But modern sciences are marked by suspicion or aversion to purpose and teleology. The genetic programme, code, and informational metaphors help to bridge this gap. According to the metaphor, control, agency, executive function resides in the power of genes (which power can be wholly disclosed by studying the mechanisms of gene function). But control of development and inheritance – agency – lies within organisms. Increasingly, as Keller has argued in the past, it is becoming apparent that making sense of life requires us to give organismal purpose its due.

As Keller has encouraged us to do in her writings, we must give metaphor its due as well. Here, we do so happily. Even if it were possible to dispense with metaphors in science, no doubt our scientific worldview would be poorer without them. In surprising ways, metaphors often guide us into worlds not yet known. And who wants a world without surprises? Yet, as Keller and the contributors here have shown, we must also collectively reflect on the subtle ways in which the entanglements between language, scientific research, and cultural beliefs carry us forward – and sometimes hold us back.