







Four decades ago, linguists and philosophers George Lakoff and Mark Johnson published an influential book on the nature of metaphors. In Metaphors We Live By they argued that abstract thought is mostly metaphorical (having a literal core extended by mutually inconsistent metaphors and incomplete without them), that metaphors are fundamentally conceptual (while metaphorical language is secondary), and that metaphorical thought is ubiquitous, unavoidable, largely unconscious, and grounded on everyday life. Despite the popular acclaim of the book and its impact across academic disciplines, their claims met with resistance as they challenged objectivist views of meaning and language.Two decades ago, in the afterword of the updated edition of their classic book, the authors insisted on some persistent fallacies that contribute to a false view of what metaphors are and do. Especially relevant to scientists is the mistaken idea that metaphors are just "a matter of words", namely, a way of speaking that can be shielded from rational thinking, and ultimately innocuous to it. Such a fallacy is related to the belief that metaphors are mainly "a matter of definition". The conversation stopper "it's just semantics" is simply stupid since, after all, meaning is all that matters. Furthermore, Lakoff and Johnson remarked that metaphors are natural phenomena, not mere arbitrary historical contingencies or cultural constructions; they are grounded in the very bodily nature of our daily cognitive pursuits.Today, neuroscientists remain ensnared in disputes about whether brains are computers or not. In fact, the very Research Topic to which this piece belongs shows that the debate about the appropriateness of the brain-computer analogy, and its pervasive metaphorical use, is alive and ticking. Of course, some will deem the exchange as useless while others will insist that, until we figure out what we mean by what we say, a proper scientific discussion is defective or cannot even start. Whatever the position, it is interesting to note that linguists' concerns about the neural basis of metaphorical thought mirror those of neuroscientists about the metaphorical basis of neural jargon.Here, rather than poking the brain-as-computer blister again, I would like to walk through a very brief history of the metaphorical brain, and then mention an alternative class of metaphors that could offer fresh, offbeat, and even rebel perspectives on our conception of that mushy little thing called the brain.Brains have been many things indeed. Four centuries ago, Descartes suggested that cerebral hydraulic automata produced behavior by powering animal spirits thought the nerves. Nicolas Steno then cast the brain as a machine. Fountain metaphors gave way to clocks. Leibniz objected: entering a brain as one enters a mill would reveal mechanical parts only, but nothing mind-like whatsoever. In the time of Shelley's Frankenstein, Galvani and Volta explored the role of electricity in animal bodies. Nerves turned into wires and brains into telegraphs. Neural plasticity rendered the analogy obsolete. Cajal preferred natural images (trees, gardens, forests). For Darwin, thought was a "secretion" of the brain. It was soon postulated that all animals, including humans, were conscious machines. Machines would become animal-like too (electric dogs, clockwork beetles, three-wheeled moths). With the theoretical articulation and empirical implementation of feedback loops, the line between biology and technology thinned. Pitts and McCulloch's "neural network" further blurred the distinction between natural and artificial ones. The metaphor was turned inside-out: computers were like brains, then brains became computers. With information theory in place, code-breaking percolated from the war to the lab. Physicalists stopped worshipping matter and turned into information, a notion as pervasive as confounded. The brain has, thenceforth, been revered as the preeminent computational organ. Despite considerable progress and obvious updates of the metaphor du jour (networks, the internet, and so on), we have not gone further.The less told story of the idea of the brain involves other images. A remarkably forgotten one, inspired by a fascinating physical technique, is the hologram. The brain would be a holographic information device, modulating waves within a holographic field. Any part could contain the whole, allowing for non-local memory storage -the elusive engram would be lost and found. Alternative metaphors do not need to be sophisticated to be conceptually juicy. Consider a prism, whereby light is reflected (perception as virtual action) and refracted (embodied affect). Blue would not be created in the prism but selected from the incoming beam. Similarly, revisiting the telephone switchboard analogy renders the brain as the organ of "attention to life" (as Henri Bergson put it), and whose main role is to receive, delay and conduit movement, selecting external images rather than producing them. From this perspective, brains are more like radios than VR headsets. The so-called resonant brain theory and the phenomenon of representational drift are kindred contemporary ideas. This whole class of metaphors leads us back to William James foundational (and forgotten) distinction between brain function as productive versus permissive. That the brain is a "reducing valve", in Aldous Huxley's words, is an intriguing hypothesis that could be investigated within the current renaissance of psychedelic research. Brains would not create thoughts but filter them. There are other metaphors neuroscientists could live by.It is also helpful to note that English is not the only language spoken by human beings. The word computer is not universally translated as such. A computer is still a "computer" in Italian, Portuguese, or German. But in Catalan we say ordinador, which we took from the French ordinateur. The same happens in Spanish with ordenador (except outside of Spain, where people say computadora). The story behind the choice of words is peculiar: In 1955, the IBM marketing team in France decided against branding their new product too similarly to existing "calculators". As a better (and shorter) name for "the new electronic (programable) machine intended for information processing", they decided on ordinatrice électronique. Finally ordinateur settled as a trademark, percolating to current language. Brains are thus pictured in the image of a person who organizes or who computes, depending on the country. We can remain lost in translation.Like fish in the sea, we often fail to notice the entrenched metaphors we swim in. Brains aren't really any of those, and yet treating brains as such can provide valuable insights unless one does not erect one's favorite image into an idol. As Lewontin's quote of Wiener and Rosenblueth puts it, "the price of metaphor is eternal vigilance". A metaphorical monoculture is a burden rather than a blessing. All metaphors are wrong (when literally taken), but all metaphors are useful (when kept in their local domain of application). Screwdrivers are handy, but not to eat soup. Entertaining other ways to conceive what brains are and what they do is not only valuable but necessary. Moreover, in the light of Lakoff and Johnson's thesis, honoring the metaphorical nature of much of our scientific thinking frees us from a shamefaced exercise of imagination and allows us to deliberately explore the many-sided nature of the brain. In the bigger picture, Metaphors We Live By was a reaction to the tendency within the analytical tradition to demote metaphors as either meaningless or simply pragmatic vectors to literal meanings. The thesis of the book was discordant with those provinces of the Western cannon that ascribe an immaculate purity to concepts, but also with the skeptic relativism of postmodernist doctrines. As we think, we live. The metaphors we use shape what we can and cannot see under our microscopes and in the real world. When it comes to understanding human mental life, the study of metaphors is complementary to the study of brains themselves.
Why mental metaphors do not help us understand chatbot mistakes
The function of chatbots like OpenAI’s ChatGPT is based on detecting probabilistic patterns in the training data. This makes them vulnerable to generating factual mistakes in their outputs. Recently, it has become commonplace in philosophical, scientific, and popular discourses to capture such mistakes by metaphors that draw on discourses about the human mind. The two most popular metaphors at present are hallucinating and bullshitting. In this paper, we review, discuss, and criticise these mental metaphors. By applying conceptual metaphor theory, we provide numerous reasons why they do not succeed in providing us with a better understanding of factual chatbot mistakes. We conclude by calling for justifications of the epistemic feasibility and fruitfulness of the metaphors at issue. Furthermore, we raise the question what would be lost if we stopped trying to capture factual chatbot mistakes by mental metaphors.
Metaphor and the Political War
Metaphor and the Political War
Metaphor, Morality, and Politics
We may not always know it, but we think in metaphor. A large proportion of our most commonplace thoughts make use of an extensive, but unconscious, system of metaphorical concepts, that is, concepts from a typically concrete realm of thought that are used to comprehend another, completely different domain. Such concepts are often reflected in everyday language, but their most dramatic effect comes in ordinary reasoning. Because so much of our social and political reasoning makes use of this system of metaphorical concepts, any adequate appreciation of even the most mundane social and political thought requires an understanding of this system. But unless one knows that the system exists, one may miss it altogether and be mystified by its effects.
Many Minds: Seven metaphors for AI
If you wanted a petri dish for understanding metaphors—how they emerge and evolve and jostle with each other—it would be hard to do better than the world of AI. We talk about AI systems variously as coaches or co-pilots, little genies or alien intelligences. Some researchers claim that AIs "grow," that they're entering their phase of "adolescence." Critics deride AI products as slop and dismiss LLMs as a kind of autocomplete on steroids. What's behind these different characterizations? Which ones are accurate and which are unfair? And are our metaphors mostly colorful rhetoric or do they matter? Are they shaping how we understand, adopt, and ultimately regulate these new technologies? My guest today is . Melanie is a computer scientist and Professor at the Santa Fe Institute. She is the author of the book, and she writes a by the same name. This episode is a bit of a companion to with Steve Flusberg. In that episode, Steve and I attempted a kind of crash course on metaphor and the human mind. Here, Melanie and I sit down for more of an extended case study: how metaphors are guiding, galvanizing, and maybe deceiving us in the contested realm of AI discourse. We unpack seven of the most widely used metaphors in this space. We consider how these metaphors are shaping not only our everyday understandings of AI, but also law and policy. We also talk about the metaphor and analogy capabilities of AI itself. Can these systems reason abstractly in the way that humans can? Along the way, Melanie and I touch on: AI-generated poetry, anthropomorphism, the original sin of AI research, the myth of Narcissus, psychometric testing and its pitfalls, metaphors for AI that are a bit hard to spot, and the question of whether an AI has ever come up with a decent analogy for itself. Longtime fans of the show will know that we've had Melanie on the show . We invited her back, not only because she's thought about metaphor and analogy in AI discourse for decades, but because she's a voice of calm insight in an area that’s increasingly awash in hype and polemic. Longtime fans of the show may also note that we are now celebrating our 6th birthday at Many Minds. That's right, the show launched in February 2020. If you'd like to support us as we recognize this milestone, you can leave us a rating or a review, recommend us to a friend, or give us a shout out on social media. Your support is always appreciated. Without further ado, on to my conversation with Dr. Melanie Mitchell. Enjoy! Notes 3:30 – For an overview of Douglas Hofstadter’s work on analogy, see . 8:00 – Much of our discussion in this interview draws on Dr. Mitchell’s piece on the in Science magazine. 13:30 – For earlier discussions of anthropomorphism on the show, see our earlier episodes and . 16:00 – See for the original discussion of LLMs as “stochastic parrots.” 17:00 – See for the original discussion of ChatGPT as a “blurry jpeg.” 18:30 – See for the original discussion of LLMs as role players. 22:00 – See for one use of the “LLMs as crowds” metaphor. See also a discussion of this metaphor (and other metaphors for AI) . 25:00 – For one discussion of AI as a “cultural technology” by Alison Gopnik and colleagues, see . For a more recent discussion of the same metaphor by Henry Farrell, Alison Gopnik and others, see . 27:00 – For the podcast series on intelligence that Dr. Mitchell co-hosted for the Santa Fe Institute, see . 28:00 – See for an influential formulation of the idea that AI is an “alien intelligence.” 29:00 – For philosopher Shannon Vallor’s book about AI as “mirror,” see . 31:00 – For the recent study on users’ metaphors for AI systems, see . 33:00 – For more on the rise of social AI, see our earlier episode . 38:00 – For more on what AI researchers might learn from developmental and comparative psychologists, see Dr. Mitchell’s (summarizing her keynote at NeurIPs). 42:00 – For more on the ARC (Abstraction and Reasoning Corpus) and the research that Dr. Mitchell and colleagues have been doing with it, see and . 48:30 – For the study on humans' preference for AI-generated poetry, see . 50:30 – For Brigitte Nerlich’s documentation and discussion of various metaphors for AI (including AI’s metaphors for itself), see . Recommendations , by Shannon Vallor ‘,’ by Murray Shanahan (!) et al. ‘,’ by Henry Farrell et al. Many Minds is a project of the , which is made possible by a generous grant from the John Templeton Foundation to Indiana University. The show is hosted and produced by , with help from Assistant Producer and with creative support from DISI Directors Erica Cartmill and Jacob Foster. Our artwork is by . Subscribe to Many Minds on Apple, Stitcher, Spotify, Pocket Casts, Google Play, or wherever you listen to podcasts. You can also now subscribe to the Many Minds newsletter ! We welcome your comments, questions, and suggestions. Feel free to email us at: manymindspodcast@gmail.com. For updates about the show, visit or follow us on Bluesky ().
Position: Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces!
Intermediate token generation (ITG), where a model produces output before the solution, has become a standard method to improve the performance of language models on reasoning tasks. These intermediate tokens have been called \say{reasoning traces} or even \say{thinking traces} -- implicitly anthropomorphizing the traces, and implying that these traces resemble steps a human might take when solving a challenging problem, and as such can provide an interpretable window into the operation of the model's thinking process to the end user. In this position paper, we present evidence that this anthropomorphization isn't a harmless metaphor, and instead is quite dangerous -- it confuses the nature of these models and how to use them effectively, and leads to questionable research. We call on the community to avoid such anthropomorphization of intermediate tokens.

Position: Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces!
Intermediate token generation (ITG), where a model produces output before the solution, has become a standard method to improve the performance of language models on reasoning tasks. These intermediate tokens have been called \say{reasoning traces} or even \say{thinking traces} -- implicitly anthropomorphizing the traces, and implying that these traces resemble steps a human might take when solving a challenging problem, and as such can provide an interpretable window into the operation of the model's thinking process to the end user. In this position paper, we present evidence that this anthropomorphization isn't a harmless metaphor, and instead is quite dangerous -- it confuses the nature of these models and how to use them effectively, and leads to questionable research. We call on the community to avoid such anthropomorphization of intermediate tokens.

Sense-making reconsidered: large language models and the blind spot of embodied cognition
Large Language Models (LLMs) demonstrate a kind of linguistic competence that theories of embodied and enactive cognition have long deemed impossible for systems lacking the meaningful perspective of a living being, i.e., the capacity for sense-making. Facing up to this unexpected technological development requires confronting what I propose to call the “AI dilemma”: either frontier LLMs are capable of sense-making despite lacking biological embodiment, or the kind of linguistic competence they exhibit does not necessarily require sense-making. In their chapter on cognition, Frank, Thompson, and Gleiser (2024) maintain that no AI system comes close to realizing relevance, a position that derives much of its motivation from past practical failures. However, frontier LLMs have effectively overcome Dreyfus’ commonsense knowledge problem, such that their dismissal as categorically mindless risks undermining Frank et al.’s central claim that human cognition is deeply intertwined with lived experience. I therefore argue in favor of the alternative side of the AI dilemma: human-level linguistic competence of LLMs should be recognized as a novel non‑biological form of sense‑making, based on a technologically‑mediated embodiment whose enabling properties are in need of further theoretical analysis. This reorientation invites enactive theory to clarify which aspects of sense-making may be universal and which aspects are specifically contingent on organic life, thereby advancing its conceptual framework in dialogue with contemporary AI.
Don't treat surface-level metaphors about AI as if they describe fundamental reality
Debates about whether AI is a “tool” or whether it’s wrong to “anthropomorphize” it often seem confused about what reality is made of

(PDF) Review of Philosophy in the flesh: the embodied mind and its challenge to western thought by George Lakoff and Mark Johnson. Basic Books 1999.
PDF | An abstract is not available. | Find, read and cite all the research you need on ResearchGate

We Have Always Been Action TheoristsToward a Critical Theory of Language for the Era of “Large Language Models”
Scholars of literature and culture understandably place themselves among the world’s premiere experts on matters of language. But they also know that fields like linguistics and communication have their own ways of studying how people express themselves through speech and written media. A key difference concerns the theories and methodologies...

Contra Literacy-Laundering: Mechanistic Critical AI Literacy
Critical discourse on large language models (LLMs) has bifurcated between epistemic dismissal that invokes some form of the stochastic parrot metaphor to puncture hype, and pragmatic accommodation that treats LLM capability improvements as grounds for updating the critique. We argue that both misdiagnose the problem as the issue is not whether or not LLMs work, but what kind of working is happening and at whose cost. Drawing on meta-theoretical frameworks of cognitive science, feminist labor analysis and critical pedagogy, we propose a conceptual reorientation. We develop this claim through registers of (i) the cognitive, examining what is forfeited when statistical pattern-matching substitutes for the iterative, grounded processes that constitute thinking; (ii) the pedagogical, examining how “AI literacy” as currently deployed is itself a symptom of the confusion it purports to address; and (iii) the political, examining how the infrastructure framing of AI naturalizes asymmetric labor displacement, particularly of feminized cognitive and reproductive work. The stochastic parrot, deployed with mechanistic precision rather than mere rhetorical convenience, specifies what is forfeited when cognitive labor is delegated, who bears the cost, and why a literacy adequate to this moment must begin from the epistemology of those most harmed by the systems it describes. We conclude with underlining that critical AI literacy, which this paper embodies an instance of, is the only sensible way forward.
The brain is a computer is a brain: neuroscience's internal debate and the social significance of the Computational Metaphor
The Computational Metaphor, comparing the brain to the computer and vice versa, is the most prominent metaphor in neuroscience and artificial intelligence (AI). Its appropriateness is highly debated in both fields, particularly with regards to whether it is useful for the advancement of science and technology. Considerably less attention, however, has been devoted to how the Computational Metaphor is used outside of the lab, and particularly how it may shape society's interactions with AI. As such, recently publicized concerns over AI's role in perpetuating racism, genderism, and ableism suggest that the term "artificial intelligence" is misplaced, and that a new lexicon is needed to describe these computational systems. Thus, there is an essential question about the Computational Metaphor that is rarely asked by neuroscientists: whom does it help and whom does it harm? This essay invites the neuroscience community to consider the social implications of the field's most controversial metaphor.

Wittgenstein Explains Why We Always Misunderstand One Another on the Internet
The best class I took in college was on the philosopher Ludwig Wittgenstein. Until that point, I had avoided philosophy of language as simply being too...

“Shutdowns Are Like You’re Stuck on the Blue Screen of Death”: A Metaphor Analysis of Autistic Shutdowns
Background: Autistic adults commonly experience shutdowns, which are distressing events widely discussed within Autistic communities. Yet, few studies to date have explored firsthand accounts of shutdown experiences. In this study, we aimed to identify figurative language that Autistic adults use to describe shutdowns, thereby broadening the understanding of and providing language for this complex phenomenon. Methods: We used a participatory research approach to conduct a secondary analysis of qualitative data from two prior studies on shutdowns in Autistic adults. Primary studies used asynchronous interviews and a survey, with a total sample size of 86 participants. We analyzed a corpus of 87 typed pages using the Pragglejaz Group’s metaphor identification procedure to identify figurative language that Autistic adults used to describe shutdowns. Results: Autistic adults in this sample used six metaphors to explain shutdowns, including the following: being frozen, a computer crash, going inside myself, when I can’t keep up, survival mode, and playing a role. Discussion: These metaphors help to externalize the internal experience of shutdowns. Participants described shutdowns as a state of being stuck, often involving physical and/or vocal immobilization. They frequently described shutdowns as a response to perceived threats, connecting them to a sense of survival and self-protection. Metaphors provide a rich context for understanding and talking about shutdowns. Findings also highlight how Autistic adults organically construct metaphors to articulate their experiences. Future studies could use metaphor analysis to explore other affective experiences of Autistic individuals. Community Brief Why is this an important issue? Many Autistic people experience shutdowns, but there is not much research on what shutdowns feel like. Studies that have looked at shutdowns found they can be hard to explain, especially to people who have not had a shutdown before. People often describe shutdowns as feeling frozen or stuck. We need to better understand shutdowns from the perspectives of people who experience them. What was the purpose of this study? We wanted to understand how Autistic adults describe shutdowns. Metaphors can help explain an experience by comparing it with something else. Our goal was to find out what metaphors Autistic adults use to describe shutdowns and if those metaphors can help explain what a shutdown feels like. What did the researchers do? In two earlier studies, we asked 86 Autistic adults about their experiences with shutdowns. Some answered questions in interviews, while others filled out surveys. Using these data, we used a method called “metaphor identification procedure,” meaning we looked for words or phrases where participants used metaphors or other figurative language to compare their experience with a shutdown with something else. Once we identified these metaphors, we grouped them together to find metaphors that summarized the experience for participants. What were the results of the study? We found six common metaphors that participants used to compare shutdowns with something else to help explain their experiences. Participants compared shutdowns with the following: being frozen, a computer crash, going inside myself, when I can’t keep up, survival mode, and playing a role. What do these findings add to what was already known? Few studies have looked at what it feels like to have a shutdown for Autistic people. Our findings add a rich description of what shutdowns feel like and add Autistic voices to the literature. What are potential weaknesses in the study? Most participants in this study were White, so we are not sure if these findings are relevant for people of other racial backgrounds. Since people needed internet access to take part in this study, we may also be missing the perspectives of people who do not have access to a computer or internet. How will these findings help Autistic adults now or in the future? These findings give non-Autistic people a better way of understanding what shutdowns are like for Autistic people. This knowledge can guide future research and help clinicians better support Autistic people who experience shutdowns.

“Shutdowns Are Like You’re Stuck on the Blue Screen of Death”: A Metaphor Analysis of Autistic Shutdowns
Background: Autistic adults commonly experience shutdowns, which are distressing events widely discussed within Autistic communities. Yet, few studies to date have explored firsthand accounts of shutdown experiences. In this study, we aimed to identify figurative language that Autistic adults use to describe shutdowns, thereby broadening the understanding of and providing language for this complex phenomenon. Methods: We used a participatory research approach to conduct a secondary analysis of qualitative data from two prior studies on shutdowns in Autistic adults. Primary studies used asynchronous interviews and a survey, with a total sample size of 86 participants. We analyzed a corpus of 87 typed pages using the Pragglejaz Group’s metaphor identification procedure to identify figurative language that Autistic adults used to describe shutdowns. Results: Autistic adults in this sample used six metaphors to explain shutdowns, including the following: being frozen, a computer crash, going inside myself, when I can’t keep up, survival mode, and playing a role. Discussion: These metaphors help to externalize the internal experience of shutdowns. Participants described shutdowns as a state of being stuck, often involving physical and/or vocal immobilization. They frequently described shutdowns as a response to perceived threats, connecting them to a sense of survival and self-protection. Metaphors provide a rich context for understanding and talking about shutdowns. Findings also highlight how Autistic adults organically construct metaphors to articulate their experiences. Future studies could use metaphor analysis to explore other affective experiences of Autistic individuals. Community Brief Why is this an important issue? Many Autistic people experience shutdowns, but there is not much research on what shutdowns feel like. Studies that have looked at shutdowns found they can be hard to explain, especially to people who have not had a shutdown before. People often describe shutdowns as feeling frozen or stuck. We need to better understand shutdowns from the perspectives of people who experience them. What was the purpose of this study? We wanted to understand how Autistic adults describe shutdowns. Metaphors can help explain an experience by comparing it with something else. Our goal was to find out what metaphors Autistic adults use to describe shutdowns and if those metaphors can help explain what a shutdown feels like. What did the researchers do? In two earlier studies, we asked 86 Autistic adults about their experiences with shutdowns. Some answered questions in interviews, while others filled out surveys. Using these data, we used a method called “metaphor identification procedure,” meaning we looked for words or phrases where participants used metaphors or other figurative language to compare their experience with a shutdown with something else. Once we identified these metaphors, we grouped them together to find metaphors that summarized the experience for participants. What were the results of the study? We found six common metaphors that participants used to compare shutdowns with something else to help explain their experiences. Participants compared shutdowns with the following: being frozen, a computer crash, going inside myself, when I can’t keep up, survival mode, and playing a role. What do these findings add to what was already known? Few studies have looked at what it feels like to have a shutdown for Autistic people. Our findings add a rich description of what shutdowns feel like and add Autistic voices to the literature. What are potential weaknesses in the study? Most participants in this study were White, so we are not sure if these findings are relevant for people of other racial backgrounds. Since people needed internet access to take part in this study, we may also be missing the perspectives of people who do not have access to a computer or internet. How will these findings help Autistic adults now or in the future? These findings give non-Autistic people a better way of understanding what shutdowns are like for Autistic people. This knowledge can guide future research and help clinicians better support Autistic people who experience shutdowns.
