• Here is a thought experiment: what happens when the cost of imitating a person reaches zero? Assume that we can create synthetic media that looks like X, speaks in X voice with X’s mannerisms etc etc — and deploy that in any channel or persistently available across all channels for anyone who wants to communicate with X. Depending on the fidelity of the model underpinning that synthetic media you would then never know if you are interacting with the person or the model – and let’s, for the sake of our thought experiment, say that we reach 99% fidelity in 85% of the possible communication contexts.

    Would this matter?

    Say I want to ask you a question, or suggest we go for dinner. I send you a quick message in WhatsApp and you reply immediately, suggesting another date and time and we agree on that. Do I care that it was not you who agreed? Probably not. Or say my children get a video call from me as I am travelling and they chat through briefly what is up, and the things they work on in school etc — and it was really a model doing that while I was out having said dinner with you. Does this matter? We probably feel it does. Yet – if they do not know? Does it matter then?

    It still feels like it does, and so what we end up with is this feeling that there is an ethics of attention. There are some things – children – that we are ethically required to pay attention to, and some other things – like friends wishing to calendar things with us – where we are not ethically obliged to be paying attention at all.

    But we need to qualify that, because we do pay attention even when it is through the model. At some point we will catch up and see that we are booked for dinner. At some point the model will summarise the cal with the kids and give us a sense of what’s going on. It is authentic, synchronous attention that we care about. There is a norm here – that we pay authentic, synchronous attention to some things – that we need to think about.

    My best guess is that authentic, synchronous attention is required in far fewer cases than we think. The transition to a world where we accept that someone is just paying “extended” attention to us through different tools and models might be quite quick – as long as commitments and decisions made in that extended attention still matter.

    How much can attention be extended? Could you imagine a corporation where everyone reports to the CEO, and that the CEO uses their extended attention to do 1:1s with everyone? This seems to hinge on the question of how much authentic, synchronous attention means for the people you work with – and how compressed those interactions can become. At the end of the day you – the synchronous, authentic attention – need to be able to monitor all of the interactions that you have engaged in, in your extended attention.

    So what is the highest possible or optimal compression rate for your daily interactions? What would be lost if you worked this way – either as a manager or a report?

    There are several different possible views here: one is that only synchronous, authenticated attention matters: this is how you share agency and purpose and trust and ideas. Leadership can never be compressed in any way. Another is that 99% of all interactions in the work place could be replaced by compressed formats, and that compression rates are really high – perhaps as high as 99%. The boring truth is always somewhere in the middle, of course, but it is an interesting question as to where you feel you are currently – in your day job.

    How much of your interactions could be compressed to a paragraph of text, say? If you gained the ability to catch up with another 100 people? And if the model could also highlight items that it knew were of special interest to you through-out the organisation you work in?

    Would you trade 10 1:1s with authentic synchronous attention with 100 1:1s compressed into three bullet points, with the important bits highlighted and decisions required outlined and specified?

    We know we should answer “no” to that question, because there seems to be a lack of respect for others to respond “yes”. The idea that we can be compressed, that interactions with us can be compressed, seems unethical. It flies in the face of some kind of means / end-distinction. It seems disrespectful and bordering on psychopathic to suggest that artificial interactions can replace real interactions.

    So, what if you could just add the 100 at low or no cost? And get them and the synchronous, authentic attention? Well, we would have to concede that those extra 100 1:1s would have some value right? Then it is a question about alternative costs – what else could we do if we could compress interactions in our daily life?

    It has been true for a long time that our mind is really extended into our tools and environments and friends and social networks. It now seems likely that our attention also will be extended, solving the Simon problem: with a wealth of information comes a poverty of attention. New technologies that create extended attention are, however, going to require that we think hard about the ethics and economics of presence.

    +
  • Speculation. If we assume, for the sake of argument (it is a big assumption), that human and artificial intelligence both are neural networks and that if trained on the same data, will both produce the same intelligence, we still are left with the question of whether humans and machines can be trained on the same data sets.

    Humans have bodies, exist in ecosystems and complex biological contexts and this, all of this, is by necessity a part of the data that we train on. Machines just train on language. images and other things that can be captured digitally. Sure, we could also capture all the possible movements of a body and deploy sensors to start to try to collect a data set that would mirror what it means to be embodied in the world but we don’t have an inkling of an idea of how much data that would be or exactly how it would look, or how to wire those sensors to capture embodiment in some sense.

    That would suggest that the data set that humans train on is radically different from that of machines, at least for now.

    But is that true? We could also argue that embodiment, evolutionary context and embeddedness are all captured in language. This argument would amount to an argument that says that language contains a compressed version of the whole of the evolutionary and biological experience – and that metaphors code for embodied existence, for example. If this is the case, language is all you need to replicate intelligence (all other assumptions here equal, including that we are also neural networks).

    The question of the equivalence between the data sets L (language) and B (embodied existence) asks a fundamental question about the nature of language, and one that is intriguing to explore. We can cut it in a multitude of different ways and try to figure out if the embodied existence we lead can be captured in other ways – but at some point this will be relevant to the recreation or mimicry of human intelligence.

    There are arguments here to be made about limits – such as: there is no way to collect data that corresponds to the evolutionary experience and history without rerunning evolution at the pace of evolution, an argument about the incompressibility of human experience as biological beings, and hence about the impossibility of a replication of human intelligence.

    All of this is very speculative, of course, but there is something here worth exploring more, a set of philosophical problems and models that may matter. Note to self: language as coding human experience effectively and efficiently would be a semiotic field of exploration – the idea of a fourth function of the Peircean sign.

    +
  • Here is an argument that I have been noodling on lately. I am not sure I agree entirely with it, but I think it is worth considering.

    It is easy to believe that you are a machine, and it gets easier the more complex the machine is. If the machine can also mimic some part of our behaviour it becomes almost impossible not to fall into the metaphorical trap and assume that mind being as machine means that mind is a machine.

    The shift from as to is is a small slip of the mind, but tempting for us as we are always trying to understand ourselves.

    Yet, we are not machines, not even very complex ones. We are human beings.

    This is not a grand statement of human exceptionalism (we should be wary of those as well), but rather a simple scientific fact. We are a part of in incompressible algorithm, evolution, that unfolds in a network of ecosystems. The machines we compare with are not.

    Now, it is easy to lose sight of this when we speak of things like “artificial intelligence”, since this, and many other concepts like it, come with a silent third term – “human”. When we say “artificial intelligence”, well, we hear “artificial human intelligence” and so we get stuck in the metaphorical trap again.

    What has emerged in AI today are extra-biological intelligences – but this does not mean that they are less interesting. On the contrary, they are stranger and more wondrous than we think. How come something that is so unlike us can behave so much like us? That is the question – not if we are mere machines.

    And this matters for everything – philosophy, ethics, policy, economics and our overall predictions for this technology.

  • One increasingly complex problem within utilitarianism is that it puts such emphasis on the individual. The idea that utility is individually felt and assessed may miss something important about utility – the fact that it is a concept that is deeply relational. Or put differently: if you were the last human on Earth you would not experience any utility or happiness generally. You would not be unhappy either, but merely incomplete.

    Now, where this goes deeply wrong is where we then revert to saying that utility is only possible to construct across an entire population – so that the population experiences a sum total utility of some kind and that utility is what we should maximise. This view also seems deeply, obviously wrong. A collective cannot experience anything at a certain size – but we miss that there are many intermittent states that we should explore more closely.

    A family, a group of friends, an organisation – or even a party! Utility is produced in the groups where there is interaction between the members at a certain level of complexity. Utility, then is an emergent phenomenon and some kinds of utility can only be produce in groups of certain sizes: just like you cannot play a symphony with only three people.

    So, the kind of happiness produced in a football stadium is different from the happiness you feel on an evening when you settle in to have dinner with your partner and children – but both are a kind of utility.

    A network producing musical utility

    Now, you may say, surely you can organise in some kind of value order? You value the dinner more than the soccer match? Well, no – that is where I actually think that we go down the wrong path again. Utility categories are incommensurable and not comparable – and what we look for is some kind of portfolio combination of different incommensurable utilities. This is why the repugnant conclusion is such a sophism — it is a version of the argument that Socrates pokes fun at in Euthydemus. If you are a father and all children have a father – then you are a father of all the children – in its crudest form.

    This is also why I like the notion of “partiality” that Tyler Cowen introduces in his talk on EA here. He makes that point that the people in the repugnant conclusion – subsisting but still by mere addition producing more utility than a few happy people – are a different species. And inter-species utility comparisons are simply meaningless when driven too far. (Does this mean that I do not care about the happiness of dogs? Of course not – I think there are some kinds of happiness or utility that you can only experience together with a dog. Again utility, like happiness, is not individual.)

    This brings me back to an issue that increasingly is bothering me – and that is the emphasis on the individual that we live with. Many philosophers have noted that the idea that we are unique atoms is simply false, yet so much current thinking seems stuck in thinking of people as isolated entities. And when that view is criticised it is contrasted with the idea that there is no individuality.

    The only way out of this it seems to me is to recognise that individual is a network concept – as is happiness and utility. It is a concept that is produced by different-sized networks. These cannot be too big, nor too small – and we need to understand the role of different network formations in policy and politics, as well as in philosophy, much better.

    All in all this is perhaps just a version of Wittgenstein’s private language argument: we cannot define our own words and give them meaning, because meaning is produced in shared language. Our happiness is shared, our utility is shared and ultimately we are shared in language as well — but really exploring the consequences and implications of that argument at different level reveals a fundamentally different way to think – a mental model – of concepts that have been hopelessly distorted by the gravity lens of the atomic individual.

    +
  • Tyler Cowen makes a series of excellent points in a recent post where he muses over the value of large model hallucinations. He notes that he would not like for these systems to stop hallucinating, since the hallucinations have value in representing something. What this something is, he suggests, could be our statistical average view of the world. Where a hallucination is factually wrong, it reveals a deeper pattern of ‘wrongness’ in our overall understanding of the world. Maybe, then, hallucinations can allow us to map areas of our understanding that are candidates for research, improvement and factual discovery?

    ‘Hallucination’ is not merely a large models phenomenon. People make things up too — a lot of the time, actually. You have surely done so as well at some point, when you did not know the answer to something – and you made, say, ‘an educated guess’. The educated guess is simply a statistical prediction given the knowledge you have trained on, and so exploring hallucinations will be a lot like exploring the nature of guessing.

    A real question here is how guessing incorporates more than linguistic knowledge – and if our guessing is better or worse than that of language models. Can we guess better if we are embodied? Guessing might be a much more important mode of cognition than we have allowed for – and may provide a really helpful mental model for thinking about artificial intelligence; not as an omniscient being, but a different guesser.

    Our default position has been to assume that what these models do is that they predict the next word or set of words in any situation. This opens the possibility of thinking more carefully about the relationship between guessing and predicting — and where we do one and where we do another. In a closed setting, such as a game, we predict – and in an open setting, or open world, we guess – or something such. There is a need for some conceptual clarity here.

    What are some ideas we could explore? Maybe the below.

    • A guess is a prediction with a certain probability range.
    • A guess is a prediction but without a probability assignment at all – and where no such assignment makes sense. I.e. guesses are about uncertainty and predictions about probability.
    • A guess is very different from a prediction in that it serves as a basis for our actions as it is made – we do not evaluate guesses as we evaluate predictions. (Somewhat unclear – what is the key we are getting at here? That guesses are reflected in actions?)
    • A guess is never about a single fact, but about a complex system.
    • A guess is about the unfolding of a narrative, a prediction is about an isolated proposition.
    • A guess is revealed prediction (as in when we say, thoughtfully, “I guess I believe he did not really mean it…”) through introspection.

    There is more to be done here — linking up the statements we make about the future in ways that help create a grammar of the future and how we relate to it in different ways. This grammar now changes with the advent of more powerful artificial intelligence, perhaps?

    + ,
  • It is a common mistake to think that modern methods of production are superior to what has come before, or that we are, in our time, at the pinnacle of knowledge. A lot of knowledge has been lost or forgotten over the years, and rediscovering old knowledge can be as powerful as producing entirely new knowledge.

    An interesting example is Roman concrete. Researchers have found that it has excellent durability – much greater than current concrete – but have struggled understanding how that can be. In a recent report it turns out that the Romans employed a method that created an anti-fragile and self-healing kind of concrete through what the researchers call “hot mixing”.

    The result of this method is a concrete that cracks in a specified way. The cracks that develop are then such that if they fill with water they heal. It is worth pausing to think about the ingenuity of this design: to predict that the material will crack, design the way the material fails and ensure that it fails in ways that can heal — that is a way of thinking that we can learn a lot from.

    The conscious design of failure into strength is not just a great method for making concrete – it is a mental model that we can use to explore different other things: everything from large language models to cars. And this mental model is a re-discovery, not a new discovery.

    The example of Roman concrete, then, suggests an interesting possibility: we should not just speak of innovation, but of re-invention of old models of construction, building, doing science and thinking. The careful study of history and earlier accomplishments can be as fruitful as plunging into the new.

    The history of science is ripe with ideas that have fallen by the wayside just to be rediscovered and become essential later. Reading deeply into the history of our subjects of interest, and not reading history as an account of failure leading to today’s insights, but rather as a source of ideas worth re-thinking, is an undervalued research method.

    The field of artificial intelligence has its own share of these examples. Neural networks were long thought dead ends, but proved to be enormously powerful tools for building machine learning systems. But what other things have been abandoned along the way that deserve re-discovery?

    Science practice might do well to assign close study of what is considered a dead end on the path of discovery. Most of science is not a labyrinth with a single path to a center, but more of a maze, with many different paths to different centers. Doubling back in a maze and examining the forking paths more deeply can yield insights that are as deep as any that are found on the way ahead. Science-as-a-maze is a mental model that also allows us to escape the idea that progress is linear – which it clearly is not.

    Scientific progress through remembering what was forgotten presents real opportunities, and arguably means that we should spend more time on understading the road not taken in many scientific fields. And maybe, just maybe, we will find curious insights that have been lost over time.

    +
  • What role does writing play in politics? Do we care about intellectuals penning long form essays about the future of the state or about official policy documents setting out frames and ideas for how to approach political issues?

    Historically the answer has been a very clear ‘yes!’. The role writing played in the shaping of the 19th century politics seems hard to overestimate. The writings of Marx, Mill and others informed and underpinned political changes in a multitude of ways. The Federalist papers were fundamental in shaping the US political landscape and as communism evolved it evolved in a terrifying mix of violence and writing. Sometimes the writing preceded the violence, sometimes it justified it – but the writing was always there.

    Political power wars were fought with texts surprisingly often. In Ezra Vogel’s brilliant study of the multiple rises and falls of Deng Xiaoping we find a story of different memos, documents, texts, opeds and other texts that shaped the political landscape in China, and continues to do so.

    Texts were not just words, they were ideology – but where do we see this today?

    We do not have to like that ideology, to be clear, and we do not even have to like the idea of ideology to ask this question. It is interesting to study just because of the consequences for politics overall. So, let’s think about the different options – what is it that we really want to say here? There are a few possible hypotheses. We could start with something really broad, like this:

    A. Our politics have become post-ideological. Texts no longer are political instruments, and they lack political interpretations. We have lost – forgotten – how to interpret texts as truly ideological, and what remains is an ability to read texts as political emotive.

    This is a very broad statement, but it is interesting to ponder. If we have lost the ability to read ideologically, to interpret texts ideologically, then something fundamental has changed in the political field as such. It would mean that there is no real ideological writing available to us to effect political change anymore – politics have become not just post-ideological but post-textual.

    If we want to sketch out why this happened, we could suggest that the rise of social media, the explosion of texts and opinions all have devalued the written word to the point where it no longer has the ability to achieve political saliency. With a wealth of text comes a poverty of textual value – and texts no longer sway us in the way they used to. Our capability to write something that means something has faded away.

    This is interesting in other ways – we could also argue that what we see here is the thinning of meaning overall. If we wanted to we could connect this back to the overall problem of meaning in a world without religion, and suggest that we have lost the ability to mean something because we have no value anchors in our language anymore. God has left our language, and so how can we really mean anything then? This is a version of Dostoevsky’s dictum that if God is dead, then anything is permitted – but it takes that observation and twists it: if God is dead, then we cannot mean anything anymore.

    As texts lost their religious salience, their political and ideological saliency faded as well – because the latter was dependent on either being in support of or opposition to the first.

    That would be a somewhat conservative take on the strong statement of our hypothesis – and you could easily poke holes in it by suggesting that there was not a lot of god in Marx – but maybe the comeback then would be that Marx was only possible as Marx in opposition to other values in society – that ideology requires friction and a general agonistic structure in language?

    Well, let’s leave that particular formulation and look at a more narrow one.

    B. Texts matter less in politics, because mass media and the Internet now compete with the text in more engaging ways. Ideology remains, but its main vehicle now is video and other social media formats. Ideology has shifted vehicles.

    This more narrow version of the hypothesis almost seems to make too much sense: of course texts lose some value when there is readily available media that are much more engaging! That seems like a self-evident truth. But the question then remains – is there a video or channel that is to the political debate what the Federalist papers were to the US political awakening? Is there a tweet that can compete with the Communist manifesto?

    You could argue that this is asking the wrong question – because the wealth of new media has not just shifted the vehicle of ideology, but the very nature of it as well. Ideology no longer is monological, it is dialogical. What shifts a society today is the interaction between the individual and multiple points of engagement – twitter accounts, video channels, facebook pages – and society really does change in that dialogue.

    This idea – of dialogical ideologies – could then draw on Russian literary theorist Mikhail Bakhtin’s writings about dialogical novels and formats. Bakhtin’s view was that all living was dialogue and that meaning only exists in dialogue, there is no meaning in the single monological statement. He gave, as an example of the dialogical style, Dostoevsky’s novels and suggested that these were the first modern novels because there was no single locus of meaning in them – his novels are poly-centric in the sense that you are equally introduced to all the people in them and forced to choose your stance in relation to them. The text does not tell you who they are, you have to engage in the text dialogically to find out who they are to you.

    Dialogical ideology, then, is an open conversation where the need for individual texts has disappeared. There is no need for the federalist papers, because there is a wealth of dialogues to engage in – and these dialogues are much more democratic and open than the discussions about texts in the past. Dialogical ideology is ideology democratised – and the reason we no longer see politics shaped by memos and letters and texts is that politics now is shaped in poly-centric dialogues.

    We do not live in a post-ideological world, we live in a dialogical ideological world where ideology is constantly negotiated in a multitude of media and with a multitude of actors.

    Is this better or worse? This is a complicated question, because it depends on the model of politics that we employ. If we argue that politics is essentially forces acting on the individual, then politics have, at a very minimum, become more complex. But if we think that we are all politically accountable, then the question of better or worse comes down to our own competence to deal with a dialogical formation of our shared ideology.

    Are we citizens enough to assume the responsibility and accountability that follows from the dialogisation of ideology?

    Which one of these do I think is true? I am not sure. I sort of hope it is the second, that ideology is still here but that it requires that we step up as citizens to ensure that we do not fall prey to propaganda and misinformation. Because the ability of propaganda to shape politics has remained strong through-out time. Vogel’s book is filled with examples of how propaganda texts shaped politics in ways that benefitted the authoritarian governments – and the loss of textual ideology is also in some ways the loss of efficient uni-centric propaganda.

    The challenge is that we now have poly-centric, dialogical propaganda and we do not quite know how to deal with it – and this in turn suggests that there is an intimate relationship between propaganda and ideology.

    +
  • The philosophy of questions is an on-going interest of mine. After having co-authored a book on questions – in Swedish, here – I have continued to keep notes and think about questions and how they relate to other subjects I am interested in. One of the things that recently struck me was that questions are a form of attention management, a way to share attention. When we ask a question, we direct the attention of others towards what we are attending to as well. This means that questions are never just acquisition of information, but rather consist in sharing attention. The information produced or acquired is secondary to that shared attention.

    This, in turn, also suggests that the way we think about questions and answers is wrong. We tend to think that an answer is directly related to a question, but it is produced as an effect of the sharing of attention that the question demands. This is why answers are a sort of shared seeing – when I give you an answer, I share what I see, or know, and you then can calibrate from that answer to a new question if you want to. Or something like that.

    The issue of power emerges quickly when we think about questions this way. Asking a question is to demand not just your attention, but also that you direct it in a certain way and spend time on what I am asking about. Questions consume attention in order to produce answers, and the right to question implies a kind of power.

    There are many different forms of questioning power, it seems. The child that asks is not wielding the same power as the interrogation officer – but both are exercising some kind of power. In the first case, it is attention willingly given and in the second attention coerced. When we ask ourselves questions we force our attention in ways that help us shape, rather than find, questions. Maybe this is why it is so hard to genuinely ask ourselves questions – it is like when you use a coin to make a decision: if you have decided to do X if the coin comes up tails, and it comes up heads and you immediately feel resistance, well, then you know your question was not open, but connected to a deeper answer already felt.

    The practice of writing dialogues – then – is a way to uncover the answers already present in ourselves, though the conscious excavation of them by attending to ourselves.

    +
  • Time to travel back to the city, sort out of a few things and then back to London. The trip back to Stockholm is 6 hours by car. I used to dread that, but have come to quite like the opportunity to think, listen to music and audiobooks as well as watch the landscape change from winter and snow to grey and rainy. The temperature here, in Storhogna, is -19 degrees Celsius. It is supposed to be +3 degrees in Stockholm.

    A trip like this would not have been possible – in the alloted time – just 100 years ago. The cars, infrastructure and safety required simply did not exist – and so we are now travelling in ways that are new. How does that affect us? That is possible to travel within a day such distances? The case gets even more interesting when we speak of air travel – since 6 hours by air will get me all the way to, say, Dubai. It seems that it should change the way we think and perceive the world.

    The most common guess is that it shrinks the world. That we perceive the world to be smaller, because it has become more accessible. There is some truth to this, but what we perceive to be smaller is not the world, it is a set of fragments that we can travel through. The world is what exists in between our destinations, and it does not shrink as much as gets compressed. The world becomes more compact, not smaller, and certain points become bigger, the destinations grow and become larger and the in-between is compressed to spaces we transition.

    We can imagine the world a bit like a novel, and our mode of travel abbreviates it. It can be accessed faster, at a lower resolution and with a lot cut out – and only a main plot remains.

    That is another possibility: that our mode of travel reduces the world to a smaller set of narratives that condenses the human story into a story about the main destinations. The ski resort and the capital are the protagonists of this story, and everything in between is obscured and cut out.

    Our mode of travel, then, edits the world into a shorter story.

    I don’t think this is bad, in itself. But it suggests that there should be other modes of travel that either are more like wandering – without destinations – or much more destination dense. Travel on foot, perhaps. Was it Walter Benjamin who made the point that just reading a text is like flying across it, and that if you really want to understand it you should translate it? That translating the text is like traveling across it on foot? I seem to remember a phrase like that, and if it is not true – well, then it still is interesting to think about what it would look like to travel in a way that is more like translating the distance into understanding, rather than abbreviate it.

    Travel as translation, abbreviation, editing. And then of course the notion of different kinds of travel. Benjamin, again, comes to mind – his last journey, running from evil, but ultimately ending his life. The reason for our travel transforms it from mere transport – into pilgrimage, escapes, journeys, tourism or expeditions.

    Why we travel and how we travel speaks volumes of us as a civilization.

    +
  • In The Primacy of Doubt: From climate change to quantum physics, how the science of uncertainty can help predict and understand our chaotic world (2022) Tim Palmer, weather mathematician and physicist, explores a set of issues related to uncertainty. It is a great read, even if it gets technical in places, because it tracks the author’s own questions and research issues. As such the book reads as a research journal in places, with personal remarks on things like consciousness, decision making and the question of religion and spirituality.

    Palmer starts from weather, and the problems of predicting weather that has faced humanity for a long time. He chronicles the move from historical patterns as the basis of prediction to complex grid models and ensemble forecasting – showing how the best way to deal with uncertainty is to represent it well in our reports and research. The practice of showing projections with intervals of certainty is just one of many examples of what he considers should be much more widely spread practices, and it is hard not to agree with him.

    It has become a tired cliche to say that a book is interdisciplinary, but Palmer’s book is that – or perhaps something better: it is a book that ignores the disciplines and explores a set of problems as if they make up their own discipline: the study of uncertainty at all levels, from quantum physics to the economy. Maybe discipline agnostic studies are more valuable than the so-called inter-disciplinary work that many seem to be paying lip service to – ignoring the disciplines rather than trying to navigate them is probably a better framing.

    Palmer certainly does so – moving from weather prediction to the question of the ultimate nature of gravity (Palmer suspects it is not a force of the same kind as other fundamental forces of physics – and that there are no gravitons), he aims to show how uncertainty at all levels is key to understanding complex systems. He makes a convincing case for understanding uncertainty, and chaos, much better – and in many ways this book is the natural heir to Gleick’s work on Chaos – and perhaps best read with its predecessor.

    It is a book to digest and come back to; there is much in here that is thought provoking – and many mental models get renovated in useful ways. Just one example, that really made sense to me: the idea of energy efficient rationality. Palmer notes that what we usually term biases are really just energy efficient ways of sort of solving problems in approximate ways where we do not want to expend the energy to get to the deeper answer. This, in turn, is eminently rational from an evolutionary perspective – since it minimizes the amount of energy we spend on problems that we perceived to be constructed or artificial.

    The only thing psychology has access to study in the lab, with students, is energy efficient rationality, and all other forms of rationality must be inferred in ways that are perhaps better suited to the methods of someone like Gary Klein.

    Palmer also suggests that there is a real and interesting role for noise to play in cognition and computing – and here I suspect that he is onto something quite important, as low voltage smaller circuits might become noisier, and if we see that as a strength and harness it we might be better off than if we try to minimize the noise.

    Finally, it will be interesting to see what happens with the reception of Palmer’s book since he discusses climate change in a rather open way – attempting to seek out strengths and weaknesses in what he terms climate maximalist and minimalist theories. It seems clear that Palmer believes that anthropogenic climate change is a real thing and has potentially disastrous consequences – but he also notes that climate models are weak in many respects (they miss things like cloud cover negative feedback loops, he argues) and so there is a risk that he is read as a climate skeptic (which seems such a weird term, as he notes, since no one is skeptical about there being a climate?).

    That would be a huge mistake – since there is a lot to learn from his discussion on climate change; not least in the discussion about cost / loss ratios and where to balance out. It seems safe to say that optimizing for energy efficiency, while acknowledging that no civilization reduced its absolute consumption of energy without perishing, is a stance that should be explored more.

    All in all: 7/10 – and worth a re-read.

    +