See blog post by Cristina Leston-Bandeira and me on the The Constitution Unit site
Are digital technologies making politics impossible? This was the question of this year's Nine Dots Prize. Here are my thoughts reflecting this question.
The term politics originates in the Greek words politikos, meaning “of, or pertaining to, the polis, the city-state unit of citizens” and politika as used by Aristotle in his book of the same name to denote “affairs of city” or “affairs of state”, given in ancient Greek cities were the governmental units. Politics are thus activities of governance pertaining to a society and its citizens. But a society can only be governed well if it is understood. And understanding requires sufficient information. Social systems in particular are characterized by high uncertainty. In order to cope with and limit uncertainty so as to be able to make sensible decisions, information is required. Information is the resolution of uncertainty. The key for governance is therefore sufficient information about the society and its members. Political action under uncertainty relies entirely on information and information processing that leads to understanding. Without information politics is essentially impossible, as it would result in unintelligent, random and reactionary action, or in activities not pertaining to the society and its citizens.
Since politics is performed pertaining to society and therefore to citizens, who constitute a society, it also requires communication, the two-way transmission of information with as little noise as possible. Politics involves the communication between citizens and political incumbents, as well as the communication among political incumbents. In particular, citizens have to be able to communicate their concerns and preferences to political incumbents. Citizens have also to be able to communicate with each other to form interest groups. Political incumbents have to be able to communicate their decisions and reasoning to citizens. And political incumbents have to communicate with each other to evaluate, to argue, to negotiate and to find compromises.
Digital technologies are information and communication technologies. They facilitate the effective gathering, processing and analyzing of information and they enable the two-way transmission of information, thus communication, on an unprecedented scale. The more processes are digitized, the more information about societies is available in the form of vast amounts of increasingly rich data. Social systems are information systems and digital technologies equip us with tools to handle and shape these information systems.
Unquestionably, humans have engaged in information gathering, processing and communication as long as humans have existed, not only since the invention of digital technologies. Human history is also a history of continuing improvement of information collection, processing and communication. However, throughout much of human history the information was sparse, communication was inefficient and in consequence politics was rather limited. Information was gathered and processed to facilitate administration and/or tyranny, but the information was incomplete, highly standardized and mostly parsimonious given limited human capacities in information processing. Information moreover served mostly the purpose to control and exploit rather than to truly understand and then to build. An excellent example of this is the Domesday Book, commissioned by William the Conqueror in 1086. The book contained information on 13,418 English settlements and provided William with insight into the financial (taxable values) and military resources available to him (cf. http://www.nationalarchives.gov.uk/domesday/).
With the rise of democracies and modern autocracies in more recent history, politics in terms of governance started to thrive supported by an optimized bureaucratic administration and advances in media and various connecting technologies (e.g. transport, electricity, etc.). And yet politics remained confined, now more than in the past, because the existing governance and administration structures and mechanisms predate the digital revolution. With rising complexity of societies these outdated bureaucratic structures have become increasingly dysfunctional, unable to respond adequately to unruly social realities (Helbing & Pournaras 2015). And so now is the time to realize the full potential of politics with the help of digital technologies. With digital technologies as they exist today and as they are likely to evolve in future humanity has for the first time tools that truly allow for comprehensive collection, processing and transmission of information and for citizens to make their preferences, feelings and opinions known. With these technologies intelligent democracy is possible. Future democracies are inevitably digital democracies. However, equally possible is intelligent autocracy. Future autocracies will be digital too.
And how do digital technologies make democratic and autocratic politics possible? I will first explain what kind of democratic politics is made possible by digital technologies and discuss then what kind of autocratic politics is facilitated by these technologies.
Future governance is data-driven governance. Only with real-time monitoring, that is real-time information collation and analysis, can governments react to our world of acceleration, where dangerous diseases can travel across the world within a day, as the last H1N1 (human influenza) pandemic in 2009 for instance showed (Helbing et al. 2015). Only real-time monitoring will allow governments to coordinate a timely, evidence-based response. Already systems (e.g. the EU-funded European Media Monitor or the Alphabet-funded GDELT Project) exist that monitor the Internet to issue nearly real-time alerts of disease outbreaks, as well as outbreaks of violent conflicts and other social, economic and political events that may similarly be contagious as new complexity-science based research shows (ibid.)
Information-nurtured politics can now be highly effective because not only can it make use of mainstream information but can easily take into account the full range of information, including information at the margins, which, accumulated, may carry significant weight. Recent research (Mann & Helbing 2017) has shown that the diversity of information, including minority information is key to finding optimal solutions and digital technologies can help to find and make use of minority information. However this research also shows that minority information is only useful if there is an incentive for gathering correct information (ibid.). Intelligent politics thus relies on accurate information and as much digital technologies facilitate the unprecedented gathering and processing of information, they facilitate on an unprecedented scale the manipulation and distortion of information. Thus, the key that makes politics possible, information, can also make politics impossible if turned into misinformation.
But even accurate information is not yet knowledge; it is not yet understanding. And here again digital technologies can help. Governments that drown in oceans of unprocessed information will not act more intelligently than governments with only limited information. In the hands of skillful analysts digital technologies however also provide powerful tools for information processing, analyzing and modeling to create knowledge. Understanding can then arise when we bring the different knowledge bits together, compare them, contrast them, converge them, reflect and challenge them, discuss them. Understanding often is based in genuine exchange, critical review and collaboration and digital technologies facilitate such exchange and collaboration again on an unprecedented scale. In the political realm, however, this would require first overcoming rivalries and mistrust, as well as bureaucratic hurdles often maintained in political and administrative institutions. Once this is accomplished, collaborative processes of understanding could be even aided by artificial intelligence (AI) assistant systems that could point out to us our biases, misjudgments and blind spots. Digital systems that correct human error do already exist in various fields from aviation and space operations to health and safety (Strauch 2004) and they could be included in politics, not to replace human political decision-making and acting but to help humans navigating the increasingly complex world and making intelligent decisions that serve the public good.
The exchange that is central to establishing understanding is also the key to (re-)linking politics to citizens. For the first time in human history a communication from many to many is a reality, giving voice to the millions of citizens that politics must relate to. Minority voices have a chance to be heard and for the first time we can appreciate the diversity of voices! As mentioned earlier, research shows that allowing minority voices to be heard and thus increasing the diversity of information increases the probability of finding an optimal solution to a given problem. Minority voices give society new information and are therefore invaluable. Theoretically, digital technologies could facilitate politics previously impossible, a politics where the political incumbents are constantly in touch with the citizens (Coleman & Blumler 2009). This does not mean we have to replace representative democracy with a direct democracy. It also does not mean that political incumbents should now blindly follow popular opinions, or the loudest opinions. Far-sighted politicians would continue to act based on best, intelligent judgment (cf. Edmund Burke in Bohn 1854) but they can make use of new technologies to better inform their judgment by the variety of citizen voices and information. Of course that would require governments and representatives to find ways to process the massive citizen input in a meaningful way and further technological developments will be necessary here, but astonishing progress is being constantly made in the field of natural language processing (Cambria & White 2014). New digital technologies promise that every voice can now be heard and this leads inevitably to expectations and a sense of entitlement. We therefore need fair procedures that give consideration to all the diverse voices and then extract a set of alternative compromises. We can now build such democratic procedures supported by digital technologies. And attempts are already made to build such systems. LiquidFeedback for instance is an open source project for interactive and transparent democratic decision making that encourages collaborative development of ideas, programs and projects and sophisticated voting on them (http://liquidfeedback.org/). Such intelligent systems implement fair deliberation procedures and could assist collective decision making for the common good, preventing unproductive, manipulative and unconstructive power politics. Politics can now re-ensure its capability to make collectively binding decisions, were the binding is not simply enforced but respected because everyone’s voice was given fair consideration.
Through AI systems we could even give those a voice, who do not have a voice currently: ecological systems, non-human creatures as well as future human generations. Their constantly admonishing voices could be built into our political decision making processes and politics would emancipate itself from myopia. Such systems could permanently monitor the state of the world we are living in, the oceans, the soil, the forests, the atmosphere, and the millions of species on our planet. Sensor technology, satellite imaging technology, measuring instruments attached to small robots, drones or tracking devices distributed around the world and connected to an Internet of Things as well as involving people as citizen scientists who report on the state of the nature around them, all this could create a network of constantly, flowing, real-time information that could be processed and analyzed by AI systems (e.g. Kehl et al. 2012), who would then translate the results into “voices” that would communicate to citizens and to political incumbents on behalf of these ecological systems and non-human species. Analytic, intelligent systems could also create future projections from the existing rich data about what life would be like for future human generations under various scenarios (e.g. Palmer & Smith 2014) and again translate these analyses into “voices” on behalf of future generations. These “voices” could protest against proposed policies that may impair nature’s and/or future generations’ wellbeing. They could appoint human delegates to vote on their behalf in political elections and to sue governments or companies for failing them. We could create a parliament of things (Latour 1993) where these “voices” are represented or the “voices” could act as AI representatives with the right to vote on governmental policies and legislative proposals. Many scenarios are imaginable when it comes to the details; the point is these AI “voices” would make far-sighted politics possible. Of course the AI algorithms creating these “voices” would have to be fully transparent and open-source, to make them trustworthy and to prevent their abuse for certain agendas. We do already allow algorithms to make trades on the stock market – and those algorithms are not transparent at all – so why not allow transparent algorithms to get involved in politics?
However, as mentioned earlier, digital technologies not only make democratic politics possible, they equally make intelligent autocracy politics possible. The information technologies now available could allow total surveillance and with total surveillance comes the power of total control in accordance with Foucault’s panopticon (Foucault 1995), with its all-present, invasive, but not necessarily visible disciplinary power. Permanent latent coercion could be established through automated monitoring and punishment systems that record and sanction every deviant behavior. Digital citizen rights groups like the Electronic Frontier Foundation justifiably point out that new digital technologies can make the world of Nineteen Eighty-Four (Orwell 1949) a reality. And the revelations of Edward Snowden in 2014 about the NSA and GCSQ surveillance activities show us that even democratic countries like the US or UK are starting to approach this dystopia.
In fact, the new tyranny may not even be perceived as tyranny, it can be disguised as a system permanently concerned with the wellbeing of its citizens, pretending benevolence, simulating dialogue and accountability, keeping citizens in comfortable ignorance, distracting them with entertainment, keeping them busy with continuous competition and self-optimization obsession while creating scapegoats against whom suppressed fear and despair turned to hate can be directed. Surveillance on a scale as facilitated by digital technologies would also allow for intelligent manipulation, totalitarian behavioral targeting, where the manipulation in the fashion of “soft” nudging would be tailored to the target individual. As Baudrillard (1981) has thought us, once we are living in a world of simulations easily created and optimized with new digital technologies, whoever has control of the representations of the world that we mistake for reality, has control of our reality. Only those resisting and disrupting this system would encounter the true tyranny of the system, most people would however fit in. The actual politics would happen behind the curtains and part of it would be creating and maintaining an illusion of a safe, prosperous, caring society, assuming this is what citizens wish, assuming citizens want being happy and content and not being bothered with complex problems. Digital technologies in these systems would promise people the salvation from the dictatorship of freedom and responsibility, they would promise people instant, permanent happiness and security instead of the effort, danger and uncertainty that freedom entails.
So, are digital technologies making politics impossible? No. They make true politics possible, but the question is what kind of politics do they make possible? Digital technologies can help us to revive and start a new era of enlightenment but equally they could help us to revive and start a new era of clean barbarism and unprecedented powerlessness of citizens. All of this and much more lies embedded in today’s digital technologies like in a cradle. Digital technologies are still comparatively young and malleable but the race for what they will become, a tool of emancipation, a tool of totalitarianism, something in-between has already started and the outcome of this race will determine future politics in one way or the other. It is our responsibility now to engage with these technologies and define the direction of their progress. Information and communication systems, AI systems have to be fully transparent. We do not need more technologies that make us sheep-like. What we do need are emancipatory technologies that make humans think and question, not indulge in ignorance and comfort. We have to resist our willingness to swap autonomy for comfort and reflection for satisfaction. Emancipatory technologies could start a new age of enlightenment based on rationality, empathy and wisdom. We can now start regaining our belief in our capability to solve problems long-term to solve global problems. We have now a chance to make use of digital technologies to grow a cosmopolitan awareness, awareness of our all interdependence and awareness of our planet. We will probably never reach a world order of true rationality, but with digital technologies we can come closer to it than ever before. With richer, more diverse information and communication, even if we don’t have 100% certainty we can better estimate the consequences of being wrong. But, there is no freedom without privacy. And only accurate information is useful for solving problems. Misinformation is worse than no information, because it seduces us to believe that we have reduced uncertainty in order to solve a problem while we are only increasing the chaotic state of the social system, making thus the problem worse. And communication is only meaningful if there is a recipient, if there are not only people who communicate but also people who truly listen, otherwise we may end up in a situation where only algorithms listen to us, learning how to hack us.
Baudrillard, J. (1984): Simulacra and Simulation. Ann Arbor.
Bohn, H.G. (1854): The Works of the Right Honourable Edmund Burke. Volume I. London.
Cambria, E. & B. White (2014): Jumping NLP Curves: A Review of Natural Language Processing Research. IEEE Computational Intelligence Magazine, 9(2), 48-57.
Coleman, S. & J. G. Blumler (2009): The Internet and Democratic Citizenship. Theory, Practice and Policy. Cambridge.
Foucault, M. (1995): Discipline and Punishment. New York.
Helbing, D., Brockmann, D., Chadefaux, T., Donnay, K., Blanke, U., Woolley-Meza, O., Moussaid, M., Johansson, A., Krause, J., Schutte, S. & P. Matjaz (2015): Saving Human Lives: What Complexity Science and Information Systems Can Contribute. Journal of Statistical Physics, 158: 735-781.
Helbing, D. & E. Pournaras (2015): Society: Build digital democracy. Nature, 527 (7576), 33-34.
Kehl, T. N., Todt, V., Veronex, M. R. & S. C. Cazella (2012): Amazon Rainforest Deforestation Daily Detection Tool Using Artificial Neural Networks and Satellite Images. Sustainability, 4(10), 2566-2573.
Latour, B. (1993): We have never been modern. Cambridge, MA.
Mann, R.P. & D. Helbing (2017): Optimal incentives for collective intelligence. PNAS, 114(20), 5077–5082.
Orwell, G. (1949): Nineteen Eighty-Four. London.
Palmer, P. I. & M. J. Smith (2014): Earth systems: Model human adaption to climate change. Nature, 512(7515), 365-366.
Strauch, B. (2004): Investigating Human Error: Incidents, Accidents, and Complex Systems. Aviation, Space, and Environmental Medicine, 75 (4), 372-372.
Read my and Cristina Leston-Bandeira's blog on How Twitter conversations highlight the different purposes of petitioning on Parliaments and Legislatures website.