Mihir Bellare

Our values, directions and activities are shaped by an array of surrounding cultures, from ones internal to our discipline (computer science) to ones that underly society. These powerful forces can be invisible, sometimes by design, sometimes due to their pervasiveness.

One of our goals in this class is to see the invisible, to understand and identify our cultures. We will move from broad elements of global culture to narrower elements, aiming to understand the unstated assumptions of our own discipline, be it computer science or, more narrowly, cryptography or AI.

In part we will do the latter by looking at themes popular in research, and asking questions about them. The critical thinking required here, however, does not come easily, for our education does not, on the whole, encourage it. The education system does strive to create excellence, but mostly in a narrow domain. To start to see things differently takes some courage and perspective. This is one reason we start more broadly, with works that exhibit this type of different-thinking in broader contexts. We also take a look at the educational system itself, and the assumptions that underly it. Once we have developed some different-thinking muscle, we will be able to see our own disciplines in a different way.

This is not to say that the books we plan to read are merely tools to develop our critical thinking. Their ideas and themes are of a significance that is broad and important, beyond our own work, towards the quality of life we may have and whatever form of society is to come.

Many people feel that things are somehow wrong. They see social injustices and ills, inegalitariansim, a planet threatened by climate change. They ask what they can do about this, while, laudibly, already doing much. The issues undeniably exist, but the effectiveness of different possible actions is not so easy to determine. We need to start by trying to understand two things. The first is the external, what are the forces and assumptions that shape the current culture. The second is internal, the recognition that before we can effectively change the world, we must change ourselves.

Questions?

This is what the professor, after a lecture, usually asks. It suggests a hierarchy, in which the professor has the knowledge and answers, and the job of students is to absorb it. Questions are expected to be clarifying ones. I have always been uncomfortable with this perspective. It does not encourage challenges to authority. In this class, I hope we can voice opinions, challenge authority and function more as equals.

That is not to say we should not ask questions. But your questions are not usually for me. I don't have the answers. Sometimes, nobody does, or there is no answer. That doesn't mean the question was moot. Some of the best questions have no simple answers, yet there is much value to formulating and posing them. So ask away, just don't expect a quick or easy resolution.

We will strive to be polite and respectful of the opinions of others without sacrificing honesty and depth. We hope to see the courage to discuss, rather than ban, controversial themes.

The readings

Putting Daniel Quinn's Ishmael first is to open with a shock treatment. We tend to see the world as a collection of different cultures, divided according to race, geography, history or income. Ishmael shows us a universal culture behind all these, one that remains invisible exactly because it is common to all. Yet, as Ishmael argues, it shapes us in important ways. We will, from this, gain insight into the world and ourselves, but also learn to look, not at differences, but at commonality, to see the invisible, which we will later apply to computer science.

As Rupert Read says in This Civilization is Finished, the presence of climate change is no longer something a rational person denies. The question is where it will lead. This book argues that our societies and cultures will need to change in drastic ways, or they will vanish. It is a background and perspective relative to which the rest of our discussions, about what to do with our lives and future, should be measured.

Kuhn's The Structure of Scientific Revolutions shows us how a disciplinary culture is formed and evolves. We will use Kuhn's framework to understand the current state of computer science, cryptography and AI as academic disciplines.

By now there is broad understanding that AI machine-learning algorithms can make socially inequitable decisions. Cathy O'Neil's Weapons of Math Destruction: How Big Data Increases Inequality and Threatens Democracy was one of the first to point this out, and remains a compelling, readable introduction to the problem and topic.

Shoshana Zuboff's The Age of Surveillance Capitalism: The Fight for a Human Future at the New Frontier of Power examines the surveillance economy. Meet the author, one of the first tenured women on the Harvard Business School faculty, in this interview with The Guardian.

William Deresiewicz's Excellent Sheep: The Miseducation of the American Elite and the Way to a Meaningful Life offers a critique of the educational system, particularly with regard to elite colleges. The book considers education as a granter of credentials rather than a place of genuine discourse and ideas, and the impact of the system on student life and stress. The 2020 revelations, of admissions to elite universities by bribery, only reinforce his points.

Gloria Steinem, a pioneer of the feminist movement, lives life to its fullest. Of her many inspiring books I picked Revolution from Within: A Book of Self-Esteem because self esteem is an issue often on the minds of students, and will be interesting to discuss.

Daniel Kahneman's Thinking, Fast and Slow shows us how bias impacts our thinking and choices.

Beyond these "long" reads, the readings page points to many shorter ones. Some relate to issues raised in the longer readings. Some are directed more particularly at cryptography and security.

Finally, the readings page lists a few specific lines of contemporary computer-science research that we could consider in light of the broader understanding gathered from the above. It suggests a list of questions we could ask, and answer, about them. The plan would be to arrive at a matrix that, for each line of work, answers the stated questions.

Is this class for me?

Perhaps sometimes, or even regularly, you feel some doubt. You feel things in the world, and society, are not as they should be. At times, you question your life choices, both big and small. Sometimes, you feel doubt, and it does not even seem to have a specific cause; it is just some underlying unease. This class is for such doubters. It aims to bring the doubts out in the open so that you can look at them. It won't resolve all your doubts, and it will create new doubts, but as you see things for what they are, you grow, and can contribute better to the world.

Not everyone has doubts. You may feel confident that computer science, and you, are engaged in an important and productive enterprise, that by improving technology we improve the world. If so, this class may not resonate with you. And that's fine. You don't have to take it.

"I am kind of interested in these things," you say, "but I have no time. I have more important things to do. I have research to accomplish, papers to write, talks to give. I want to be successful in my area, publish a lot and be recognized. What's wrong with that?" Nothing is wrong with that. But, I would gently suggest, you will be more successful at your goals, more productive in your use of time, and less stressed, if you do absorb some of the ideas we will see in this class. If you try to do research non-stop, you may find your productivity and effectiveness decreasing. Instead, use some time to do the readings. A book a week, not that much, and then some discussions, where you can vent, in the company of sympathetic peers. As your perspective broadens and your critical thinking develops, it will percolate into your research, in the form of different problem choices and higher-quality outcomes. Your work will be more inspired. Well, hopefully.

"I didn't finish the reading this week," you say. "Should I show up to class?" YES. Just skim the book, and you can still participate in, and benefit from, discussion.

Prevailing assumptions

If we are to take a somewhat counter-cultural stance, we should start by asking what are some prevailing assumptions, or elements, of the culture of computer science. These are not things we are explicitly told in college or graduate school, but rather things that seem, implicitly, to underly what we do. Below, the term "technology" is meant to include algorithms and theory, not just computer software and hardware. Here are a few possible such assumptions that we can discuss:

  • Technology is morally and ethically neutral.
  • Growth is good.
  • Whatever problems technology creates can be solved by more technology.
  • The faster (computers, algorithms, ...), the better.
  • Problems are solved by allocating money to them.
  • We need more research.

The role of Ego