Brian Hamilton

RSS · Archive

Albert Borgmann says in Technology and the Character of Contemporary Life that a distinguishing feature of a “device” is the “concomitance of radical variability of means and relative stability of ends.” What matters about a device is what it does, not how it does it. “The concealment and unfamiliarity of the means” is the point: a device relieves us of the burden of caring about how it works.

This sounds like the kind of thing a critic of technology would exaggerate, on the way to portraying, as Borgmann does, the thing a device produces as a commodity that “can be enjoyed as a mere end, unencumbered by means.”

But I’m realizing how much this understanding of the device is an explicit goal in computer engineering. The distinction between the abstraction and implementation is fundamental. Noam Nisan and Shimon Schocken call it a “cognitive gift” (The Elements of Computing Systems, 2nd ed.) and “the most important design principle in hardware and software construction projects.” The “abstraction-implementation paradigm helps developers manage complexity and maintain sanity.”

In practical terms:

Whenever your implementation uses a lower-level hardware or software module, you are to treat this module as an off-the-shelf, black box abstraction: all you need is the documentation of the module’s interface, describing what it can do, and off you go. You are to pay no attention whatsoever to how the module performs what its interface advertises.

We’ve got a new member on our team who’s been doing a better-than-usual job of spinning up quickly. He took it upon himself to inventory the last ~6 months of recurring alerts and start figuring out which were genuine problems that needed attention and which alerts just needed to go away. What a smart first task: it’s an almost universal pain point, so if you can make a dent you’ll improve your team’s quality of life immediately, and understanding where the service you’re working on is hitting its limits seems like a really high-leverage way to start learning the service itself.

God forbid I have to spin up on a whole new system anytime soon, but if I do, I’m going to try this tack.

I’m only just discovering that I can use VS Code directly in my browser against an existing Github repo. This is great. Hopefully this will make it a bit easier for me to post to this blog from my work computer.

Or I could just give up and sign up for Micro.blog.

Parmenides on opinion

Parmenides has a strange relationship with “the opinions of mortals.”

On the one hand, they are clearly false. “Mortals have made up their minds to name two forms, one of which they should not name, and that is where they go astray from the truth.” By “two forms” he seems mainly to mean light and dark, but it’s really the idea of duality itself he has a problem with. Dark, presumaby, should not be granted the status of being “its own thing,” when it’s really just the not-being of light. It’s nonsense anyway to talk about what does not exist, since there’s nothing to talk about. The problem with “the opinions of mortals” is that we’ve built our entire perception of the world around these dualities, sometimes even turning them into elaborate and sophisticated sciences in their own right.

And yet for some reason I don’t understand it’s important to learn these opinions. Not just important—the goddess says it is necessary. The narrator must learn the opinions of mortals alongside the truth, even though there’s no reason to trust them (οὐκ ἔνι πίστις ἀληθής).

The clearest reason we’re given is ὡς οὐ μή ποτέ τίς σε βροτῶν γνώμη παρελάσσῃ, which I’m initially tempted to read as meaning, basically, “so nobody can make you look stupid.” Maybe learning the opinions of mortals is a kind of survival tactic, a practical necessity given that most people are not privy to the kind of divine knowledge the narrator has.

But I’m not quite convinced by that cynical interpretation. The goddess also adds this cryptic tidbit:

ἀλλ᾽ ἔμπης καὶ ταῦτα μαθήσεαι, ὡς τὰ δοκοῦντα χρῆν δοκίμως εἶναι διὰ παντὸς πάντα περῶντα.

My Greek isn’t good enough for this, but it sure sounds to me like some stronger claim is being made here on behalf of the way things appear to be (τὰ δοκοῦντα). It sounds like the goddess might even be saying that the appearances are actually in some strange way true, though that contradicts everything else Parmenides says. My ignorant attempt to translate it would go something like:

But nevertheless, you’ll learn these things [the opinions of mortals] too: how it was necessary that the appearances really exist, pervading everything.

The two translations I have to look at right now are not much help. Richard McKirahan has:

But nevertheless you will learn these too–how it were right that the things that seem be reliably, being indeed, the whole of things.

What does that even mean? “How it were right that the things that seem be reliably”?

The public domain John Burnet translation has:

Yet none the less shalt thou learn these things also,–how passing right through all things one should judge the things that seem to be.

I think what Burnet is trying to get at here is that the goddess wants to give the narrator a way of understanding how mortal opinion went wrong. That fits much more easily with everything else Parmenides says, but it looks to me like a pretty willful rendering of the Greek.

And then there’s the fact that, in the opening scene, the narrator is being flown through the night sky by the daughters of the sun, to where Justice protects the gate between dark and light. It may be that this is all just the narrator’s own immersion in mortal opinion that he is learning to overcome, but that seems too easy. The duality of truth and opinion seems to be inscribed into the cosmos itself.

Maybe there are well-worn explanations for my confusions here; I haven’t read any of the literature. But to me, for now, this instability makes Parmenides much more interesting.

Ὁ Κατάσκοπος is different than most of the “easy Greek” out there. It’s not a graded reader; it doesn’t really start easy and get harder as you go. Instead, it keeps the vocabulary easy the whole time and lets the syntax be as hard as it needs to be the whole time. I wish we had more books like this. Most graded readers wait so long to introduce anything remotely complicated—even the future, much less the optative!—that it’s impossible to get any volume with it.

Two new essays on ignorance and moral reasoning

I was surprised this week to receive page proofs for not just one but two essays I had all but forgotten about. I wrote them both about two years ago, which is a long time even by the sad standards of academic publishing. But I’m very happy to see them finally coming to print, because they still feel vital.

Both essays make the same basic argument: that a certain practice of ignorance is central to moral reasoning. The heroes are Socrates, who claimed only the “peculiarly human wisdom” of knowing he knew nothing, and Augustine, who claimed that what knowledge he had was only a gift received after despairing of trying to obtain it. Both essays are working through what it means to teach moral reasoning.

The slightly earlier of the two, “AI and the Pedagogy of Ignorance,” was written for a Society for Ethics Across the Curriculum conference in October 2023. The immediate practical question in view there was whether this new-fangled thing called ChatGPT had a place in our ethics classrooms (in short: no!). Generative AI has changed in some fascinating and relevant ways since 2023, but I think the main argument holds up. It will be in the next issue of Teaching Ethics.

The other, “What are We Teaching When We Teach Moral Reasoning?,” was written for the Society of Christian Ethics conference in January 2024. That one responds to Jonathan’s Haidt’s dismissal (in The Righteous Mind) of “the rationalist delusion” by arguing that ancient “rationalists” like Socrates and Augustine have a much more subtle understanding of reason than Haidt does.

If you read either one and have thoughts, I’d love to hear from you!