Comments from a panel at reunion 2018

Topics/tags: Talks, miscellaneous, rambly

For some unknown reason [1], I was invited to participate in a faculty panel at reunion [2] entitled Pedagogy, Precision and the Porousness of Truth in the Age of Trump. This musing represents my attempt to think through the questions and to reflect a bit on what I said (as opposed to what I planned to say). You will find most of my planned text in Roman face, my commentary in italics, and the questions in bold and italics.

0. Please introduce yourself.

I forgot to plan for this one. Fortunately, I went third, so I could adapt to what my colleagues said. I’m glad I went third; I noticed that both of my colleagues introduced their scholarship, so I thought a bit about that (I didn’t do it). The following is my attempt to rewrite what I ad-libbed.

Hi. I’m Sam Rebelsky. I teach computer science. I just finished my twenty-first year at Grinnell. I feel like a bit of outlier on this panel [3]; I don’t talk much about post-truth society in my classes and I’m not quite sure why I was invited. My only guess is that I’m known to talk about social and political issues in my class. Or maybe it’s because there’s a technological component to this panel and I can serve as the token technologist. Given the deep expertise of my colleagues, you can count on me to give comparatively shallower answers to the questions.

Too many of the people who develop computing technology look like me. That leads us to a narrow focus. One of my primary goals as a computer science faculty member is to address issues of underrepresentation in my discipline. If a broader variety of people develop computing technologies, there are better opportunities to think about impact.

1. Please discuss a pedagogical and/or personal challenge of teaching students in an era where truth and falsity compete for legitimacy, and where consensus and self-censorship threaten the pillars of academic freedom of speech and thought.

I didn’t/couldn’t quite read what I had written, but it was relatively close to the notes below.

I’m not sure that I completely understood all of the parts of the question, but I’ll do my best to address what I see as some common threads in the question.

An important issue that affects my teaching is the growth of what the popular media call echo chambers; people talk and read primarily within their own ideology and are therefore seem less inclined to address the experiences, ideas, and thought processes of others with any seriousness. I’ve watched Grinnell’s own echo chamber [4] have negative effects on my students; students who enter as moderates sometimes become much more conservative because they get so frustrated at the more extreme liberal perspectives and the way that those who seem to feel otherwise about other issues are treated.

The biggest challenge I face in the classroom is how much and how I can talk about politically and socially tinged issues, particularly in a world in which faculty members regularly end up on watch lists or as the target of social media campaigns [5]. There are certainly a number of people who think that politics and social issues have no place in the CS classroom. But, well, I’m a Grinnell faculty member; I consider talking about these issues one of my roles. After all, we are to prepare our students to use their knowledge and their abilities to serve the common good.

I also know that the groups less well represented in computing are those who seem to be most interested in the ways that computing can have broader social impact [6].

At times, the simple fact that I talk about issues in my classroom has a large impact. Students tell me that they expect to discuss complex campus issues in their social science classrooms but find it surprising in the science classroom [7] so when I talk about bias-motivated incidents or sexual assault on campus they know it’s something important.

I also know that expressing my opinions in the classroom can affect students’ willingness to express their own opinions. If it’s clear that I oppose what I see Chuck Grassley [8] standing for, how comfortable will a Grassley supporter be in expressing their opinions in my classroom? And when I’m unhappy with the way my students have acted, such as when a group of students disrupted Ken Adelman ’67’s talk [9], I’m not sure how reasonable it is to express that displeasure in the classroom. I do try to make it clear that I will listen to other opinions, but the power dynamic is complex.

2. With split-second access to internet information that disrupts attention to detail and disciplined inquiry, how has teaching students now changed from 20 years ago? 10 years ago? 5 years ago?

I did not look at what I had written, but I did follow the general points of the first paragraph below. I was responding, in part, to some of the things my colleagues said before me.

I was just discussing these issues with a student the other day. There’s a question as to whether students remember less factual data because they have such easy access to it. I see power in remembering connections more than facts and in knowing ways to find information. Knowing the context of information is important, and I think that our students think of more context, rather than less.

I tried to insert an example about strawberries and migrant labor but failed. On the other hand, I was able to say something like In reading a draft of my answer to the prior question, one of my students said something like I wrote about a similar issue. Let me check to see if I remember that right. Yes, that sit. Did you know that Crane Metamarketing mentioned the Grinnell echo chamber in one of their studies?’ She might have been able to do the same thing in the pre-Internet world, but it would have been much harder and I’m not sure she would have remembered all the connections.

I included something similar to the following paragraph.

In my discipline, students really do need to know some facts without repeatedly looking them up, such as the vocabulary and grammar of the programming languages we use and the common algorithms they’ve written or read about. They also need to remember ways to approach problems. That’s always been a challenge for some students. But it strikes me that more are having difficulty memorizing the vocabulary they need in my introductory class.

I did not discuss the following paragraphs.

I do find that students do still need to be taught how to do good Web searches and how to process the information they receive. Even in my 300-level classes, I seem much better than my students at finding relevant materials when students have a question or encounter an error.

I think the interview process for internships and jobs in computer science helps remind students about the need to rely less on what they can find online and more on what they keep in their head. The typical in-person interview involves the student working at a whiteboard, trying to design a solution to a computational problem and then expressing that solution in a programming language. To be successful, students need to know and apply strategies and be able to do more than just look up (which they can’t do) or memorize. (There are ways that this process is biased; but I appreciate that it does suggest to students that the most important characteristics are what they can do, not what they can look up).

3. What insights and legacies do you hope to instill in your students that transcend the content of your courses?

I did not include the first two paragraphs. They did not seem as relevant as I thought they would. [10]

I mentioned this question to my research students this morning. One of them said I appreciate that you tell us that The ultimate outcome of a successful liberal arts education is the ability to bullshit coherently and compellingly about any subject. I suppose I should explain what I mean by that [11].

A liberal arts education should help you quickly analyze a topic from multiple perspectives, formulate some ideas about the topic, and communicate those ideas. Bullshit seems to be a reasonable term for that combination of skills. To analyze and communicate, one needs to draw upon a wide variety of fields: You may have to think about or incorporate literature; you may have to consider societal issues; you may have to think formally or interpret or use quantitative evidence; you may have to think about how you’d test a hypothesis. You also have to know something about your audience and be able to reflect on other perspectives on the issue. Now, many folks can bullshit in a few areas close to their core areas of knowledge. But liberal arts graduates not only understand a broad range of areas, they also know how to quickly get up to speed on others. That’s a laudable outcome.

I included something similar to the following paragraphs. If I recall correctly, I didn’t really look at my notes; I just riffed on the general ideas.

I try to inculcate in my students a sense of responsibility to others. I do that, in part, by modeling. I show them that they matter to me. I encourage them to support each other in the activities they do by offering them credit. And, at times, I talk about social and political issues.

As my colleagues have suggested, one of the most important things we do is to teach our students to read closely and to question what they read. One of my favorite teaching experiences happened when I debriefed with my Tutorial students. One told me, Early in the semester, you asked us to explain the structure and evidence of an author’s argument. I realized he didn’t have one. It changed the way I read.

As a computer scientist, I also have a responsibility to help students consider the risks and limitations of computing. I should teach them not only what computers can do, but the implications of what the computers can do. Too often, technologies get built without careful consideration of possible effects or even on the groups of people who the technology might effect.

4. What role do history and its legacies have in your teaching in terms of context and choice of material?

Yeah, this was a fascinating one. Does CS have a history written by the victors? I don’t think so, but I’m not sure. I certainly value alternate historical narratives; mom and Howard Zinn were close friends for many years and I still have the copy of A Peoples’ History that he gave me. But how do I talk about that in my discipline? I didn’t say exactly the following, but I followed the general arrangement of topics. I went first so that I wouldn’t have to be as deep.

Traditionally, very few computer science courses contain any discussion of history, let alone challenges to our understanding or interpretation of history. Nonetheless, I do try to include history in my courses. In part, I do so to remind students that the things we are taking as facts were, in fact, developed by humans. I also do so to remind my students that people are fallible, not least because their vision can be limited

I use as an example the binary search algorithm, which is an important part of the computer science literature. Jon Bentley, a famous computer scientist, in writing about this algorithm, first tells us that it’s an algorithm that most experienced computer programmers get wrong, and then tells us that Donald Knuth, another famous computer scientist, points out that while the first binary search was published in 1946, the first published binary search without bugs did not appear until 1962. [12]. Bentley then goes on to demonstrate techniques for writing a version you can be confident is correct. More than 30 years later, someone discovered a hidden assumption in Bentley’s algorithm that made most common implementations (which were based on Bentley’s) fail in some cases.

I also try to find ways to talk about issues of gender and race in the history of computing. The recent film, Hidden Figures, has helped a lot in those discussions.

In my upper-level CS classes, particularly our software design class, I look for cases that help students think about their ethical responsibilities. I also challenge students to identify their own.

And there you have it. The primary parts of my contributions to the panel. I did help with a few additional comments during the question and answer portion and I did talk to a few alums afterward. However, since those weren’t planned responses, I have not attempted to record any part of them here [14]. I’m told that someone drew the talk [15]. I look forward to seeing the picture.

Oh, I also found out why I got invited. If I recall correctly, the comment was The person who suggested you didn’t know much about your scholarship, but they know that all students love you. While I think I did an okay job, there are at least two problems with that statement. First, not all students love me. Some really dislike me. Second, even if lots of students love me, that says nothing about my ability to talk sensibly about complex topics, particularly topics relating to truth. Fortunately, an underlying theme was pedagogy and I care deeply about pedagogy.

[1] Okay, it’s not quite unknown. I interact a lot with the faculty member who was originally planning the panel and they appear to think highly of me (at times, more highly than I think of myself).

[2] More precisely, at Alumni College, which is part of reunion.

[3] I had a colleague from Sociology and a colleague from Education.

[4] One of my research students said: Did you know that Crane Metamarketing wrote about the Grinnell echo chamber?

[5] It appears that one of my fellow panelists was the target of such a campaign. I hadn’t known.

[6] Did I mention the awesome student who started to read over my notes. She suggested this idea.

[7] Yes, I know that not everyone considers CS a science.

[8] I had originally planned on saying Trump. But I’m pretty sure that one of my colleagues told an alum beforehand that we wouldn’t mention that President’s name.

[9] An alum came up to me afterward and suggested that the students should have asked a question that tied his excoriable behavior at Grinnell and beyond. I think it went like this How do your poor choices in social groups at Grinnell, which seem to be at heart of your inexcusable treatment of others relate to your poor decisions in selecting Donald Rumsfeld as a mentor and your actions afterward? I think they had hubris in the question, too.

[10] Okay, by the time I got to the talk, I decided not to use that opening.

[11] The explanation below is based on a prior musing.

[12] Jon Bentley. 1983. Programming pearls: Writing correct programs. Commun. ACM 26, 12 (December 1983), 1040-1045. DOI=http://dx.doi.org/10.1145/358476.358484

[14] I should record one. If I recall correctly, someone asked how we train our students to be critical thinkers. I was surprised to hear my colleagues’ answers; they seemed to suggest that our students need a lot of support in that. My sense is that our students have good foundations and are interested in being critical thinkers. I was tempted to say In this case, as in many others, I do my best to model what I want the students to learn. At least that’s what I hope that I do with some of my musings. It’s also what I try to do when I work through a complex paper with students in Tutorial (or, on occasion, in other courses).

[15] No, not Brandy Agerbeck ’96. It appears we have more than one alum who is talented at summarizing discussions visually.

Version 1.0 released 2018-06-01.