Philosophy from the Zettabyte

Interview by Richard Marshall.

Luciano Floridiis Professor of Philosophy and Ethics of Information at the University of Oxford, where he is also Director of Research and Senior Research Fellow of the Oxford Internet Institute and Governing Body Fellow of St Cross College. He is a member of Google Advisory Council on the Right to be Forgotten. His last book is The Fourth Revolution: How the Infosphere is Reshaping Human Reality(Oxford University Press, 2014). Here he broods on the history of philosophy as a sine wave, on why the philosophy of ICT is ultimately a species of the philosophy of ethics, on what information is, on a post-analytic-continental divide, on hyperhistory, on the infosphere, on the ethics and politics of information, on responsibility, privacy, Google's The Right to be Forgotten, on the quality of information and on why AI is interesting because of what it tells us about ourselves. This one's about who we want to become...

3:AM:What made you become a philosopher? Are those reasons still sustaining you or are there new reasons why you continue as a philosopher?

Luciano Floridi:I was exposed to philosophy very early. My parents are passionate about it and my father introduced me to Russell, and in Italy philosophy is taught for three years in high school. By the time I was seventeen, I had three answers to the proverbial question “when I grow up I want to be…”. An economist, to understand the human world. A mathematician, to understand the world. Or a philosopher, to understand. Since I wanted a license to study any subject in abstract terms, the choice was easy, intellectually. I followed Terence’s advice: “Homo sum, humani nihil a me alienum puto” (“I am a human being, nothing human is alien to me”). In practice, it was much harder. At some point in the nineties I was close to giving up, for what I wanted to do, namely to develop a philosophy of information, did not seem to interest any philosophy department, anywhere. Today things are very different. But my motivation has not changed. I still want to understand. I am still searching, like a teenager, for the ultimate meaning of life (there, I said it). I still wish to be a student of everything. I agree with Shakespeare that “All the world’s a stage”, but I disagree that “all the men and women merely players”. Some of us have the privilege of being in the audience and study the play. However, whatever understanding I have gained so far has come at a price. I have lost two fundamental beliefs. I have become agnostic (mind, not atheist, for I do not believe in the non existence of God either), and I have now realised, with the certainty of the heart, that I will die not knowing. I miss God deeply. And my irrecoverable uncertainty (questions without answers) hurts badly.

3:AM:You argue that there’s an information revolution going on, that it’s very important and to ignore it is not an option just as it wasn’t with the first three – the Copernican, the Darwinian and the Freudian. What is it about the zettabyte era that is causing the revolution?

LF:The history of philosophy looks a bit like a sine wave (or a roller coaster, if you prefer). It goes up and then down, up again, and then down. The ups, the crests of the wave, are the innovative periods, when we deal with philosophical problems. These are the times when philosophy is engaged with open and fundamental problems in relation to its own time. Once successful, philosophy fells in love with its own image, which is admittedly beautiful and attractive to any speculative mind. And like Narcissus it drowns, unable to leave the beauty of its reflection. The downs, the troughts of the wave, are the scholastic periods, when we deal with philosophers’ problems. I believe that the information revolution is a great opportunity to renovate philosophy and climb up again on a new crest. Academic philosophy is definitely too narcissistic today. It would be very healthy to make it look at the world, instead of itself. And the world itself is in great need of philosophical understanding and design of new ideas. We need philosophy on board while we are creating the information society and re-thinking what I like to call the human project.

But what kind of philosophy? It seems to me that it should be a philosophy engaged with the profound transformations caused by information and communication technologies (ICTs). No aspect of human life is being left untouched by ICTs: education, work, conflicts, social interactions, entertainment, politics, art, literature, cinema, law, health, business, science… it is hard to think of anything that is not being transformed or redefined by the information revolution. This means that old philosophical problems are being upgraded, think of issues about personal identity, memory, the nature of knowledge, the foundations of science, ethics, and so forth. And new philosophical problems acquire prominence: what is the nature of information? What happens to power in an information society? Can we reconcile human freedom and its predictability by smart machines? These are just a few examples among many.

Unfortunately, if you talk to philosophers, most of them are dismissive of technology. They tend to reduce it to a matter of tools and practical skills. I disagree. I guess I am not Platonic enough. The philosophy of information is not a matter of developing a philosophy of the next gadget. It is about engaging with the deep transformations caused by ICTs in how we understand the world, hence in our epistemology and metaphysics; in how we make sense of it, hence in our semantics; in how we conceptualise ourselves, and what we think we can be or become, hence in our theories of education, identity, and our philosophical anthropology; in how we interact with each other, how we manage and shape collaborative and conflicting relations, and how we may construct the society we want, hence in our socio-economic and political thinking. ICTs and the infosphere they are creating are providing the new environments in which we live and think. To use a word I introduced some time ago, they are re-ontologising our realities. Surely this is what philosophy should try to understand and help to shape properly. So, ultimately, it is a question of ethics, or, as I prerfer to put it, of e-nvironmental ethics. It is time to go back on top of the crest of the sine wave.

3:AM:You have argued that the mind doesn’t want to get information for its own sake, nor to just represent reality but rather wants it to build a model of it so it makes sense so it can defend itself from it and survive. Is this where you begin defining what you mean when you talk about the philosophy of information?

LF:It is one way of approaching it. Imagine reality in itself as a sender of messages. Reality, understood as the Big Radio, broadcasts a very wide spectrum of signals. We, humans, are able to receive some of them directly, some others indirectly. For example, the visible spectrum is the portion of the electromagnetic spectrum that is detectable by the human eye; we perceive invisible radiant energy (infrared, electromagnetic radiation with longer wavelengths than those of visible light) through technological mediations. Out of all those signals, we make sense of the sender itself. It would be utterly naïve to think that the signals are a description of the sender, yet this does not mean that they are any less real. We only have to admit that the Big Radio is not sending selfies. With two other, different analogies, we cook with some ingredients (data from the world) but the dish we obtain (information) is not a copy of the ingredients. Or, we build with some materials (data in the world), but the house we obtain (information) is not a copy of the bricks we used. Human knowledge works in this constructionist (not constructivist, mind) way, it is not mimetic, it is poietic.

The mind reacts to the non-mental (call it the noumenon, if you prefer a Kantian vocabulary, although, to be utterly honest, the dialectics I am referring to is more Fichtean) by producing a meaningful interpretation of it. Some parts of this interpretation are heavily constrained by the signals we receive. In the long run, we ask more questions to get more data, as Francis Bacon already suggested. We manipulate the data to see what further data we obtain, and all this leads to scientific theories, which are our best ways of making sense of the constraining affordances (my preferred definition of data) provided by the realities we are studying. Some other parts of this interpretation are more flexible and malleable, the constraining affordances provide much more latitude, and well-informed, rational disagreement is more difficult to resolve, think of economic policies during a financial crisis.

The mind emerges from reality by detaching and differentiating itself from it through informational layers that end up in the creation of a language and a culture, a vision of the world and a scientific explanation of it, all things that the mind calls home. It is this meaningful environment that is essential for the mind to flourish, and the mind will do anything in order to generate it, including telling stories about mythological creatures, magic energies, or anonymous market forces. There is nothing relativistic, or anti-realist (to use some common jargon) in this, in the same sense in which there is nothing relativistic or anti-realist in the dish we cooked or the house we built. Humanity has taken advantage of the signals sent by the Big Radio progressively well, this is why our knowledge works so successfully. But we do not need to embrace any naïve realism, or representational theory of knowledge, or a correspondentist theory of truth. We should think about our knowledge of the world not in terms of painting it but in terms of engineering a model of it.

3:AM:You say that you’re coming from a post-analytic-continental divide perspective. Why is this important to understanding your approach to the philosophy of information? Are you suggesting that the philosophy of information should be considered a new kind of first philosophy?

LF:It is important because I would like to see a better balance between control and power. Allow me to explain this with two caricatures.
On the one hand, analytic philosophy excels at controlling the philosophical discourse. An exact vocabulary, logic, formal distinctions, scientific information, empirical or thought experiments, mathematical formulations, statistical data, cogent and coherent arguments, a piecemeal and inferential way of discussing problems… these are all ways in which analytic philosophy can exercise a high degree of control over a philosophical topic. The “but” is represented by the risk that so much technical control may be exercised over nothing, minutiae and irrelevancies, what I called above philosophers’ problems. As John Locke once remarked, logicians keep sharpening their pens, but never write. It may get worse, if the degree of what can be controlled ends up determining the scope of what is worth investigating philosophically.

On the other hand, continental philosophy excels at enriching the philosophical discourse with powerful thoughts. An evocative vocabulary, rhetoric, scholarly references, literature, art, poetry, socio-political analyses, historical facts and interpretations, a more narrative style, existential and religious approaches to problems… these are all ways in which continental philosophy can add profound, powerful contents to a philosophical topic. The other “but” is represented by the risk that so much rich and powerful content may spill all over the place and be vague, confusing, incoherent, and sometimes downright mistaken. As a famous slogan of Pirelli (the tyre company) reminds us “power is nothing without control”. In this case too, it may get worse, if the power of the content ends up promoting irrationality and an irritated impatience towards logic, or anti-scientific views, relativism, obscurantism, and an oracular philosophy.

The best philosophy (the one you find on the crests of the sine wave) has always combined a high degree of control with very powerful ideas. And this is what I hope a post-analytic-continental divide perspective may regain. It is certainly what we need today.
As for the philosophy of information, I can only hope that it will mature into a first philosophy. Anything less and it will have failed in its task of providing us with the powerful and controlled ideas that we need to shape and make sense of the human project in the twenty-first century.

3:AM:Should we approach this philosophy metaphysically or analytically, or both?

LF:Actually, I would say neither. We need to approach it from a design perspective, which I guess cuts transversally across the analytical-metaphysical divide. Philosophy deals with open problems, that is, problems that are constrained by facts and figures but ultimately solved by neither. These are problems such that you and I could be informed, rational, and not stubborn about them and still disagree about their acceptable solutions. We move forward when we can design (not invent, not discover) ways in which they can be solved satisfactorily. Yet opting for a metaphysical approach means having the illusion that we can talk about reality in itself, without accepting any level of abstraction, that is, any level at which questions may become sensibly solvable. With an analogy, it would be as if you and I disagreed on the value of a second-hand automobile without even being willing to accept that such value must be given within a framework of considerations (financial value, historical value, emotional value, running-cost effectiveness value and so forth).

However, I suspect that opting for an analytical approach would mean denying that the same philosophical problem, precisely because it is philosophical, is open to disagreement, and hence subject to more than one solution. Even in mathematics we are used to equations that have more than one solution, infinitely many solutions, no solutions at all, or solutions that can only be approximated. Philosophical problems are not different. If we wish to find their solutions, a reasonable approach is to clarify what the purpose is that determines the level of abstraction at which the question that is being asked is actually answerable, in case agree on further constraints, and ultimately accept that there may be many solutions, some preferable to others depending on the purpose for which a level of abstraction has been privileged. Plenty of philosophers’ problems fail to be clear about all this, and become sources of endless diatribes, turning into cottage industries and scholastic monopolies.

3:AM:You use the term ‘hyperhistory’ to mark the end of old history and the start of this new era. Is this the sort of shift that fashionable ideas of ‘singularity’, ‘posthumanism’, cyberculture’, for example, are supposed to mark? Why do you find them unconvincing and that ‘the hole is way deeper’? Why do you think philosophy can help in this? Why not leave it to the scientists and IT boffins?

LF:I use “hyperhistory” to mean “more historical than before”, in the following sense. Prehistory is any stage of human social life characterised by the absence of ways to record the present for future consumption. This normally means writing, and more generally ICTs. Then we talk about history whenever individual well-being and social welfare start being related to the development of ICTs. Once individual well-being and social welfare start depending on ICTs increasingly heavily, then we can speak of hyperhistory. As a test, you know you live in a hyperhistorical society if this can be the target of a cyber attack. To put it more dramatically: in hyperhistorical societies those who live by the digit (may) die by the digit. Now, the appearance of hyperhistorical societies is related to a major shift in our philosophy and the appearance of a philosophy of information, not unlike historical events and philosophical ideas were coupled in the Enlightenment, for example. This macroscopic shift has generated attempts to explain what is happening under our eyes. We sense a deep and widespread transformation. So, fashionable ideas, such as ‘singularity’, ‘posthumanism’, ‘cyberculture’, are at worst philosophical snake oil, at best evidence of growing pains: we are confused, in search of new certainties, in need of meaningful frameworks, and so we resort to the ageless practice of telling stories, some reassuring, some scary, all fanciful. What we need to do is to develop a robust, controlled and rich philosophy. This should not be left to bizarre speculations. But it cannot be delegated to “scientists and IT boffins” either. This because they usually do not deal with open problems and with the design of the ideas necessary to answer them, with the ultimate goal of making sense and shaping the world. When they do, they are simply stepping into a philosophical debate, often rather naively. You need experts in “multisolvability”.

3:AM:What do you mean by the ‘infosphere’ and why does it mean we need to rethink our philosophy of nature? Do you think the infosphere changes the way naturalism (tricky term) needs to be thought about?

LF:Infosphere may mean many things. It can be used as a better synonym with cyberspace. “Better” because it is more inclusive as it does not presuppose an online/offline divide, something that social media have made anachronistic in the last decade (I prefer to speak of “onlife”). Philosophically and more technically, I have also used it as synonymous with reality. But not with reality in itself, for I remain rather Kantian about this point. Rather, it is the way in which we make sense of reality today. Whatever the world is itself, today, for our culture, it is becoming an infosphere. By this I mean that we have moved from a view of the world that was Newtonian and mechanistic, made of things and then properties of things and then processes in which things and their properties are involved, to an informational view of the world, in which entities are nodes arising from networks of relations, properties are special kinds of relations (for example, being tall is a unary relation, as it takes only Alice to be satisfied, whereas being taller is a binary relations as it requires both Mary and Bob to be satisfied), and processes are transitions in which networks move from one state to another. I have been guilty in the past of simplifying this view by saying that what is real is informational and vice versa, but this has to be interpreted from a constructionist perspective, according to which real has an epistemological not an ontological meaning. The sort of informational realism I defend is not metaphysical but epistemological, to be a bit more precise. Using a Kantian terminology this means that what is real phenomenologically (not noumenally) is informational and vice versa.

3:AM:You argue that politicsneeds to change too, don’t you, so that multi-agent systems can be set up to deal with global issues? Can you sketch out the issues here?

LF:Yes I do. Let me simplify as much as I can. Big problems require big agents to be solved. The big agents we have inherited from modernity are the Nations/States. In many cases, these have already evolved into multi-agent systems, think for example, of the separation of executive, legislative and judiciary powers in many democracies, the significant movement towards devolution within countries like the United Kingdom, or the federal organisation of countries like the United States. Unfortunately, Nations/States are not big enough. They are necessary but insufficient to deal with huge challenges – such as the financial crisis, global warming, health risks, terrorism, conflicts, inequality – that require more powerful agents. This is why I am moderately in favour of more robust approaches to the role of the EU, of the UN, or other similar international organisations, and of international treaties that can give humanity a better chance to rise to the global challenges it is facing today. The UN and the EU were the outcome of the Second World War. I very much hope that we will not need another tragic lesson and so much terrible suffering before we move forward in favour of the creation of some powerful multi-agent systems that may save us from our own mistakes. I am concerned that, currently, we are leaving this multi-agent systems policy in the hand of multinationals. They are the closest we have to global multi-agent systems. Nothing wrong with them, but politics should be done in terms of social agreements, rather than economic interests.

3:AM:And ethical concernsalso get shifted. One of the things that you discuss is extending the scope of the ethical environment so that artificial, digital and synthetic environments are covered. Can you set out what this means for things such as the ethics of vandalism, bioengineering and/or death?

LF:I believe we can refine our sensitivity towards what deserves, initially and in an overriding sense, some degree of respect. This is not just Nature, but Reality more inclusively. The idea is not new. Precedents of it can be found in Plotin and Spinoza, for example and, according to colleagues more knowledgeable than me, in Buddhism. I think Martin Luther King Jr. put it very insightfully when he wrote, in Rediscovering Lost Values, that “If we are to go forward, we must go back and rediscover these precious values—that all reality hinges on moral foundations and that all reality has spiritual control.” I completely agree. If one considers that nothing deserves to be proactively disrespected to begin with, then forms of vandalism, for example, become reprehensible even when they target inanimate things or artefacts that are not necessarily valuable to anyone. Vandalism – the most violent form of which is murder – impoverishes the world ontologically, decreasing its rich variety. In this sense, we can lament death as a loss of a special structure. And we can be open-minded about bioengineering as a way of adding new pages to Galileo’s book of nature. The major difficulty in all this is to counterbalance an exclusively agent-oriented position with an equally important patient-oriented one. This means structuring the ethical discourse around the question of what is good for the receiver of the action, rather than for its source. Most applied ethics in recent times belongs to this patient-oriented tradition, from environmental ethic to medical ethics, from bioethics to feminist ethics.

3:AM:Does this new situation change the link between agencyand ethics and law? We’ve seen law beginning to shift to extend legal responsibility to institutions such as banks, for example, rather than individuals working in the banks. Is this the kind of thing you’re thinking we need to get a philosophical grip on?

LF:The ethics of responsibility could make a significant step forward by engaging with at least three problems that are becoming pressing. One is distributed responsibility, the case in which the interactions between many agents and technologies, through time and different circumstances, bring about good or evil effects. Another is the distinction between moral responsibility and accountability. There are many moralactions that can be traced back to non-individual and non-human sources, from a company to a software program. And finally, but perhaps most importantly, we should free ourselves from a logic of afterlife judgement, punishment and reward, which is typical of the Christian way of thinking of good and evil actions.

3:AM:You’re advising Googleat the moment aren’t you? The big issue there is privacyisn’t it? Can you say something about how your method of abstraction might be applied in this?

LF:I am a member of Google’s Advisory Council to the Right to Be Forgotten. And yes, privacy is a major issue. Here is an elementary example of the importance of using a clear method of abstraction when discussing it. Privacy and freedom of speech are fundamental principles of any liberal and democratic society. However, if people are confused about the level of abstraction – if they use only a single agent model, for example, and apply “being fundamental” as the only criterion to evaluate the two principles – then they conclude that they must be compatible. Alice has a fundamental right to her privacy and a fundamental right to her freedom of speech and there is nothing that indicates a potential conflict. However, once we include the possibility that different agents may appeal to each principle with respect to the same contents, then Alice’s privacy can easily conflict with Bob’s freedom of speech. This is what happens with the so-called right to be forgotten, which needs to be modelled in terms of different agents’ conflicting interests in implementing two fundamental principles.

3:AM:Have you recommendations about the way forward around this issue? And is there a later impact on this – one that changes views of who we are and how we relate to others?

LF:The ruling of the European Court of Justice was a wrong step in the right direction, which is that individuals must be able to exercise more control over information that concerns them. However, I doubt that putting Google in the driving seat may be a solution (it is currently Google that decides which information is or is not de-linked, people can then appeal; and it is Google that decides who counts as a public figure to whom the ruling of the Court does not apply). It should be a legal institution that decides. Google, or any other search engine, should merely comply with the decision. Yet the decision should not be about mere accessibility, that is, about the possible removal of links from a search engine result. It should be about availability, that is, about the possibility of blocking at the source a specific bit of information. This would solve the territoriality problem, for if you block a piece of information, it does not matter which search engine tries to index it. The trouble is that this measure is a form of censorship and therefore, being a serious move, it requires taking equally seriously the harmful nature of the information in question. Unpleasantness would not count. Unfortunately, I am not sure any of the parties involved wants to get this serious. This is a pity because who we are online is now fluidly mixing with who we are off line. I mentioned already that we have begun to live onlife. The consequence is that our social selves affect who we are and who other people think we are, and this affects who we may be or become, and how we may be treated. The construction of our personal identifies onlife will be increasingly important. We better take care of the modalities though which this happens.

3:AM:A related issue is about the quality of information in our new situation. Good quality information will help us thrive it is supposed – so what do we mean by information quality? Is there a settled view and if there isn’t does this matter? What’s your position on this?

LF: The most common definition of information quality is “fit for purpose”. This helps but only partially. For it means that one needs to know the purpose and hence the level of abstraction at which the quality of the information in question is being evaluated. Ultimately, there are some parameters that can be used: accurate, complete, consistent, relevant, timely, updated, and so forth. However, the final evaluation requires an interpretation of the nature of the question which the information is supposed to answer. Asking whether some information is of good quality in itself is a bit like asking whether some food (grass? plankton? mice?) is of good quality in itself independently of the organisms that is going to eat it.

3:AM:AI was perhaps a key point for the kind of philosophical landscape you discuss, which raises the issue of the notion of whether there’s still an element of hype in all this that might not be earned. I’m thinking of the way that the hard problem of consciousness hasn’t been sorted despite films like ‘Ex Machina’ suggesting that if you build a big enough and fast enough computerized thing then it will be. Searle and Chalmers’ warnings loom over that project it seems to me and so it may be that some of the claims about the revolutionary changes aren’t as clearly present as you claim. Is this a fair pushback and does it suggest that the change is maybe not as deep as it might appear?

LF:In broad terms, I agree with Searle and Chalmers, and I have very little philosophical interest in AI, if by this we mean the real thing, that is, the non-biological reproduction of human intelligence by engineering means. This is science fiction, which I enjoy enormously, but as entertainment. However, I do find smart technologies philosophically significant. They are replacing us in an increasing number of tasks, and they can predict our choices in an increasing number of cases. They are not important in themselves. What matters is what they say about us. We are the only intelligent and free agents in the universe that I know of (although the universe is a big place, and who knows what’s happening in some of its distant corners). The special mix of consciousness and intelligence that characterises us may be a natural and yet unique phenomenon. We may be, to use a Greek word, a hapax legomenonin Galileo’s Book of Nature. This is a word that occurs only once in a text, like “Honorificabilitudinitatibus” (“the state of being able to achieve honours”), which Shakespeare uses only once, in Love's Labour's Lost. There is nothing supernatural in a hapax legomenon, it just does not occur again. If so, we better come to terms with our solitude in the universe. There are neither gods nor machines that can be blamed for our mistakes, or to which we can delegate our responsibilities. There is only us, all the way down to the end of time.

3:AM:And which five books, other than your own which we’ll all be rushing out to read, can you recommend for readers here at 3:AM to take them further into your philosophical world?

LF:If I restrain myself to five philosophical books, and I force myself to avoid the most obvious classics that readers are likely to know already, then I would recommend:
1. Gilles Deleuze and Felix Guattari, What is Philosophy?
2. Jean-Francois Lyotard, The Postmodern Condition: A Report on Knowledge
3. Herbert Simon, The Sciences of the Artificial, 3rd ed.
4. Alfred North Whitehead, Process and Reality, corrected ed.
5. Norbert Wiener, The Human Use of Human Beings: Cybernetics and Society, 2nd ed.
And if I may add six recommendations about reading, they would be:
1. Read only classics. There is no time to confront any lesser foes.
2. Choose your classics. Do not enter into other people’s fights, it is your life, you are at least entitled to choose your foes.
3. Never read a classic defenceless. A classic will deeply and irreversibly conquer your mind, so entrench yourself carefully, by digging deeply into your own thoughts, and make these strong enough to withstand the assault that will be almost irresistible.
4. Never study a classic, interpret it. After resisting its assault, counterattack as violently as you can: misinterpret a classic, steal from it, use it for your own purposes, be unfair, force it to confesses what you need to know, reduce it to something else, never show any hermeneutical mercy.
5. Never underestimate your reading list. Choose your foes judiciously. The order in which you will engage with classics will forever determine who you are. Two minds will be very different, depending on whether they wrestled with The Tempestbefore or after Faust, with Anna Kareninabefore or after Madame Bovary, with Life A User's Manual before or after Zen and the Art of Motorcycle Maintenance: An Inquiry into Values, with Waiting for the Barbariansbefore or after Waiting for Godot.
6. Reread the classics. The Leopardis not the same classic at fifteen, thirty-five, or sixty-five. The more you read your classics the more you can make peace with them. What were once foes become loyal allies, who will join you in your new battles against other unknown classics.

ABOUT THE INTERVIEWER
Richard Marshallis still biding his time.

Buy his book hereto keep him biding!