Interpretive Labor

post by Benquo · 2018-09-05T18:36:49.566Z · LW · GW · 41 comments

This is a link post for https://acesounderglass.com/2015/06/09/interpretive-labor/

Contents

41 comments

41 comments

Comments sorted by top scores.

comment by cousin_it · 2018-09-06T09:04:30.326Z · LW(p) · GW(p)

The post seems to conflate two different things:

Newbies to a field need to absorb jargon to keep up, or experts need to slow themselves down and explain it.

Here you're asked to improve your understanding.

Bullied kids are told to forgive and forget because their bully “is trying to say they’re sorry”, even after repeated cycles of faux-apologies and bullying.

Here you're asked to act low status, no understanding required.

I'm in favor of (1) but not (2).

Replies from: pktechgirl, Benquo
comment by Elizabeth (pktechgirl) · 2018-09-07T03:27:08.203Z · LW(p) · GW(p)

Yeah, in retrospect that was a mistake. They're related, but it begins the slippery slope to using "interpretive labor" to mean "all emotional labor"

-Post author

comment by Benquo · 2018-09-06T14:57:14.203Z · LW(p) · GW(p)

I agree that those are substantially different. It seems like to some extent the "interpretive labor" framework suggests that (1) is more often something like a fancy version of (2) than it might otherwise appear. (For instance, if you live under a totalitarian ideological regime such as e.g. many public schools, you can't defend yourself without speaking in the regime's language, which will be constructed to make it hard for people in certain situations to defend themselves.)

Things in the class of the courtier's reply seem relevant here (see also YVain's article on it).

Replies from: cousin_it
comment by cousin_it · 2018-09-06T15:43:16.406Z · LW(p) · GW(p)

I guess my position is closer to Yvain's than yours. The courtier is worth listening to. On the margin, people would benefit from doing more interpretive labor, not taking more offense at it.

Here's a less emotionally charged example. Imagine you're a programmer at a big company, and your manager tells you to do something stupid for business reasons that don't make sense to you. Should you say "ugh, interpretive labor"? No! Understanding the manager's reasoning - not asking tons of questions, but getting a feel for how it works - is the door to the stairs to the next level of your career. I've seen so many programmers ignoring that door for years and wondering why they stay grunts. Don't get stuck like that! Jump on every chance to do interpretive labor!

Replies from: Benquo, Benquo, Benquo
comment by Benquo · 2018-09-07T20:36:14.927Z · LW(p) · GW(p)

I've thought a while (with some help from friends) about what made this comment seem objectionable to me. I think what's bothering me about it is that the linked post is trying to describe a situation where power imbalances cause asymmetric incentives (and ability) to perform interpretive labor. It seems to me like this sort of comment subtly recasts the frame into a debate about whether people should act according to these incentives or indignantly refuse, thus conflating a descriptive claim with a policy recommendation of indignation.

This is an example of the sort of thing I was trying to point to in Model-building and scapegoating [LW · GW]; descriptions of a situation that might be a problem are read as intent to blame one class of participant, often one who seems to be benefiting. My post on actors and scribes [LW · GW] was also trying to point at this.

Replies from: cousin_it, Elo
comment by cousin_it · 2018-09-08T07:44:15.631Z · LW(p) · GW(p)

power imbalances cause asymmetric incentives (and ability) to perform interpretive labor

In my example the power imbalance is mostly an effect, not a cause. You can apply to work at a programming company as a programmer, and they will accept you if you pass the interview. Or you can - with the same name and face - apply to work as a manager, and they will accept you if you pass the interview. Or you can move from programming to management, I know a bunch of people who did that. Your position is determined by your skills.

You could say "oh, but people in lower positions are prevented from learning the skills needed for higher positions". But that's not true in my experience. Programmers aren't prevented from learning "people stuff", they know they could do it, it just makes them yawn.

Replies from: Benquo
comment by Benquo · 2018-09-09T03:23:34.410Z · LW(p) · GW(p)

OK, "cause" was too strong - correlate with.

comment by Elo · 2018-09-07T20:46:32.221Z · LW(p) · GW(p)

Would this problem be resolved by adding awareness to "which one I am". When throwing around self or external blame?

"I recognise I'm blaming everyone else for my problems"

"I recognise I'm. Blaming myself for my problems"

Replies from: jessica.liu.taylor
comment by jessicata (jessica.liu.taylor) · 2018-09-07T21:51:43.030Z · LW(p) · GW(p)

Given that the original post was not blaming anyone for anything and was interpreted as such anyway, this would not resolve the issue.

Replies from: Elo
comment by Elo · 2018-09-07T23:58:27.031Z · LW(p) · GW(p)

The original post didn't claim to have awareness of itself. So it's left as an exercise to the reader to slot themselves in and create that interpretation.

It's a different style to appeal to one side.

A different style to appeal the second side.

A different style to appeal to one side and be clear about it.

A different style to appeal to the second side and be clear about it.

A different style to appeal to neutrality.

A different style to appeal to neutrality and know it.

A different style to appeal to sidedness "universally".

A different style to appeal to sidedness "universally" and know it.

A different style to appeal to sidedness "universally" and know it and get it wrong.

A different style to appeal to sidedness "universally" and know it, and get it wrong, and know it, and show it all.

A different style to appeal to sidedness "universally" and know it, and get it wrong, and know it, and show it all. In a neutral frame while still being useful to the earlier 10 stances.

Trouble is, the further up this tree that one climbs, the cloudier it is to see clarity from gibberish.

Replies from: jessica.liu.taylor
comment by jessicata (jessica.liu.taylor) · 2018-09-08T00:16:54.828Z · LW(p) · GW(p)

I am disturbed by this comment. It seems like you are interpreting the post as doing coalitional politics and ignoring the descriptive content. The fact that descriptive statements about power dynamics automatically get interpreted as doing coalitional politics is the whole issue here.

If this goes too far then all words will be interpreted as violence and we will no longer be able to talk ourselves out of this mess because such talking would itself be interpreted as violence; everything would degenerate into an unproductive and destructive mess of zero-sum conflict. I am actually really worried about this happening and want to strongly push back against it.

Replies from: Elo
comment by Elo · 2018-09-08T07:19:37.067Z · LW(p) · GW(p)

Notice that I didn't claim to be one side or another. I didn't even claim to be critical or uncritical of the post. Or claim which level the post was on. And there are more levels above.

Also everyone is always playing on many levels. But also we trust and take charitably each other's ideas. Without that trust we really do live in zero safety world.

comment by Benquo · 2018-09-06T20:09:55.378Z · LW(p) · GW(p)

People don't face the binary choice “do more interpretive labor” and “take offense at it,” and it’s inappropriately political in this context to generically exhort people to work harder to internalize the preferences of those who have power over them.

comment by Benquo · 2018-09-06T20:07:17.196Z · LW(p) · GW(p)

Why didn’t you give this example instead?

Imagine you’re a manager at a big company, and one of your programmers is slacking on a task for motivational reasons that don’t make sense to you. Should you say “ugh, interpretive labor”? No! Checking in with your direct reports to make sure they understand the business case for what you’re asking and how it might help them advance in their careers—not asking tons of questions, but getting a feel for whether there’s shared understanding—is the door to the stairs to the next level of your career. I’ve seen so many managers ignoring that door for years and wondering why they’re stuck with a team of underperformers. Don’t get stuck like that! Jump on every chance to do interpretive labor!

Replies from: SaidAchmiz, cousin_it
comment by Said Achmiz (SaidAchmiz) · 2018-09-06T20:35:08.688Z · LW(p) · GW(p)

Why not both?

Replies from: Benquo
comment by Benquo · 2018-09-06T21:16:53.272Z · LW(p) · GW(p)

I think that's the same question for all the purposes I care about - why name exactly one example, and specifically an example of a subordinate doing interpretive labor towards their manager? This sort of selective attention is the mechanism by which asymmetric demands for interpretive labor are made.

Replies from: SaidAchmiz, Elo
comment by Said Achmiz (SaidAchmiz) · 2018-09-06T22:29:20.392Z · LW(p) · GW(p)

The question is, who are we talking to, when we write? And why?

Is the reader of a comment on Less Wrong more likely to be a manager, or more likely to be a subordinate? I’d wager that a much higher percentage of the readership fall into the category of “grunt programmer who disdains boring business stuff” than fall into the category of “manager, frustrated that his programmer underlings are slacking on tasks”.

So that’s the “who”; and as for “why”, well, if you’re a subordinate whose career is floundering, you could benefit from advice to do more “interpretive labor”; this could even be a game-changer for you. If you’re a manager, you’ll probably do fine in life without reading advice on rationalist blogs.

Furthermore and most importantly, the advice to the subordinate is good advice regardless of whether the advice to the manager is also good advice.

There’s a mode of writing, and of thinking, where one takes the “god’s-eye view”, and forgets to consider to whom one is supposed to be talking to. “Everyone should do more interpretive labor—subordinates and managers!” A fine sentiment—from the god’s-eye view. But no one chooses “everyone”’s actions; people choose their own actions only. The subordinate can choose to do more “interpretive labor”, or not. The choice before him is what it is, regardless of what we might proclaim from the god’s-eye view.

Replies from: Benquo
comment by Benquo · 2018-09-07T00:08:26.427Z · LW(p) · GW(p)

One might equally well advise workers to correct the power imbalance caused by the asymmetric interpretive labor bottleneck by unionizing, of course.

Replies from: SaidAchmiz
comment by Said Achmiz (SaidAchmiz) · 2018-09-07T00:41:01.802Z · LW(p) · GW(p)

Again, why not both?

“Stop ignoring business stuff if you want to advance career-wise, even though it doesn’t come naturally and is hard. Also, unionize.”

(This is assuming, of course, that unionizing helps at all. I take no position on this.)

Replies from: Benquo
comment by Benquo · 2018-09-07T00:56:19.194Z · LW(p) · GW(p)

No one's argued for "not both" here.

comment by Elo · 2018-09-06T21:54:11.035Z · LW(p) · GW(p)

What's that? Having high self agency and being proactive is a win state?

Why yes we say that often. Around these parts.

comment by cousin_it · 2018-09-06T20:42:27.392Z · LW(p) · GW(p)

Managers already do the bulk of interpretive labor though. And too often programmers just yawn when the conversation turns to business stuff.

Replies from: Benquo
comment by Benquo · 2018-09-06T21:18:01.466Z · LW(p) · GW(p)
Managers already do the bulk of interpretive labor though

Why should I believe this? One manager typically manages multiple people, so at first I'd expect that each employee spends much more time interpreting their manager than their manager spends interpreting them. Remember that some of the manager's time will also be spent interpreting their manager, which competes with time spent interpreting their employees.

I wonder if this is in part a representativeness heuristic problem. Because managers have to control the behavior of so many other people, they may end up spending nearly all of their time doing interpretive labor, whereas managed people have to interface with their object-level tasks much of the time. So interpretive labor is a more characteristic activity for a manager. But this is a response to the underlying dynamic where the many-to-one relationship between managers and managed makes managerial interpretive labor scarce. Interpretive labor is characteristic of managers because they have less total capacity to do it per subordinate than their subordinates do per manager, not more.

Replies from: cousin_it
comment by cousin_it · 2018-09-06T22:16:12.705Z · LW(p) · GW(p)

each employee spends much more time interpreting their manager than their manager spends interpreting them

Agree about the time ratio. But interpretive labor of managers is more efficient per time spent, because they specialize in people, while programmers specialize in computers. For example, if you want to present a new project to superiors or partners, a good manager can craft the right communication in a day, where a brilliant programmer could spin their wheels for a week and in the end the message would fall flat. The same is true for manager-programmer conversations I've seen, managers are far better at reading them and it comes from skill, not position. That's why I say programmers have more room for growth.

comment by Raemon · 2018-09-07T20:55:21.385Z · LW(p) · GW(p)

Pre-emptive Moderation Comment:

A lot of the concepts discussed in the OP (and subsequent discussion) touch on identity politics. People often come to that sort of discussion with very different assumptions and strong feelings which make it easier than normal for discussions to talk past each other, or feel that other people are talking in bad faith, etc.

I think it's important to be able to talk about this sort of thing, but politics is hard mode [LW · GW] for rational discourse. I don't have any particular comments on the existing discussion at the time I write this, but think it might be helpful if people spent a bit more effort on conversational charity than normal.

comment by Chris_Leong · 2018-09-06T00:59:12.968Z · LW(p) · GW(p)

"With a few very specific exceptions (accents, certain disabilities, ), interpretive labor flows up the gradient of status or privilege"

I think things are much more complicated in today's society, at least in Western countries. There has been a strong push back against this, which has broadened people's understandings of certain aspect of underprivileged groups (still undoubtedly quite limited). On the other hand, this has reduced many people's empathy for groups that are considered privileged.

As an example, when the government sold a private street to someone who wanted to try to extract money from the people who lived there, because they were like $300 behind on tax and they weren't properly notified, I saw so many people who were happy to see these people getting screwed over because they tended to be rich. Moving into more social justice territory, you get expressions like, "male tears" or claims that something is a "privileged" concern, which can be considered a form of negative interpretative labour; that is a refusal to interpret even straightforward claims about those who are considered "privileged".

"Speaking without thinking is much easier. Like, stupidly easy" - As Scott Alexander said, Beware Trivial Inconveniences [LW · GW]. The inevitable effect is that less things will be said, especially when these inconveniences stack. And having to keep these concerns in mind, reduces the amount of working memory available for other tasks. Further, there's a cognitive bias here as you see the person standing in front of you who is hurt, but not the person who didn't say something because it was too much effort. I'm not necessarily taking a position on trigger warnings, just pointing out that it isn't just about people being lazy.

Replies from: Benquo, Benquo
comment by Benquo · 2018-09-06T02:26:06.890Z · LW(p) · GW(p)

My understanding is that the "male tears" thing originated as a response to a perceived case of a demand for the kind of asymmetric sympathy many of the examples in the post are about, for instance:

Girls are told “snapping your bra means he likes you” and then expected to no longer be mad about it.
Bullied kids are told to forgive and forget because their bully “is trying to say they’re sorry”, even after repeated cycles of faux-apologies and bullying.
Fundamentalist husband expects his wife to know his emotions and correct for them while he actively hides the emotion from himself.
A paraphrased quote from Mothers of Invention: A woman’s house slave has run away, greatly increasing the amount of work she has to do herself. She writes in her diary “Oh, if only she could think of things from my point of view, she never would let me suffer so.”
Poor people are more empathetic than rich people.

I think it's important to distinguish asymmetric demands for interpretive labor, from complaints about such demands (although it's not always clear what's going on without a lot of context, and sometimes the latter transforms into the former in a game of telephone).

Replies from: Chris_Leong
comment by Chris_Leong · 2018-09-06T06:38:58.809Z · LW(p) · GW(p)

How this phrase originated is much less important than how it often functions in effect which is: "I have no duty to make any effort to understand you, in fact, I'm allowed to intentionally or recklessly misinterpret what you say". I guess the point is that circumventing this requires extra labour, ie. "I'm not saying men have it worse, but.."

Replies from: Benquo
comment by Benquo · 2018-09-06T21:31:55.243Z · LW(p) · GW(p)

I think overtly politicized contexts in which people are responding to a long history of asymmetric demands for interpretive labor are not the right place to start. We'd be better off focusing on mundane examples where people spend most of their time, like school or work or talking with their friends.

comment by Benquo · 2018-09-06T02:20:04.676Z · LW(p) · GW(p)

While the house example does describe something that works against rather than for the (legibly) privileged, I don't see how it's specifically about interpretive labor - can you explain more?

Replies from: Chris_Leong
comment by Chris_Leong · 2018-09-06T06:32:43.840Z · LW(p) · GW(p)

Well, when those people tried to voice their complaints they were dismissed by many commentators. So it's an example of people being unwilling to expend perform interpretive labour to emphasise with them.

Replies from: Benquo
comment by Benquo · 2018-09-06T21:21:35.596Z · LW(p) · GW(p)

There's no evidence that the house owners instead empathized with the many commentators, nor would doing so have particularly helped them.

comment by Ben Pace (Benito) · 2018-09-05T18:45:32.782Z · LW(p) · GW(p)

I didn’t know that your use of this phrase had such a clear post behind it. This is a great post, thanks for putting up the link.

comment by nBrown · 2018-09-07T10:43:38.024Z · LW(p) · GW(p)

Thanks for the writeup. The very next day after, the phrase "interpretive labour" occured naturally in my internal monologue. It's as if I'd been carrying around the concept hole / shape for quite a while.

"ahh finally, I've a way to refer to that thing." I wonder what thinking looks like before you have a handy pointer. You have to move the felt sense around all the time? A word coining seems like the creation of a central Blegg/Rube [LW · GW] node in the network.

Wonder if some folks find interpretive labour enjoyable. Currently, I enjoy stream of consciousness writing much more than writing blog posts.

For me talking to non ingroup folks can sometimes be tiring because of the need to do interpretive labour / translate out of my native brain-speak on the fly. Though some people I know don't get tired by this.

comment by totallybogus · 2018-09-07T19:33:06.721Z · LW(p) · GW(p)

There’s a whole chain of schools that teach poor, mostly minority students business social norms, by which they mean white-middle-class norms.

Are "white middle class norms" substantially different from, um, black middle class norms, hispanic middle class norms, asian middle class norms and the like? If they are, the article should perhaps hint at this, and at some relevant evidence. If they aren't, the "white" bit seems pointlessly divisive in a rather obnoxious way. Either way, you're creating quite a bit of "interpretive debt" that the reader will have to pay down via interpretive labor.

Crucial Conversations/Non-Violent Communication/John Gottman’s couples therapy books/How to Talk so Kids Will Listen and Listen so Kids Will Talk are all training for interpretive labor.

We could add the guide to How to Ask Questions the Smart Way to this list. Pithily, the "smart way" to ask a question in a technically-complex setting is the one that minimizes interpretive debt, via adopting "tell culture" norms. There are other best practices that point in a related direction, such as, in a work environment, being very clear about whether you actually understand what's being asked of you, and whether you're taking on a serious commitment to achieve it (something that plenty of people don't seem to realize as being important).

I think a large part of the anger around the concept of trigger warnings is related to interpretive labor.

I think a large part of the anger about trigger warnings - on both sides - is no longer about sensible and effective trigger warnings. Trigger warnings make sense precisely when they shift a large interpretive or emotional burden, away from the person who is least equipped to handle it.

Replies from: Benquo
comment by Benquo · 2018-09-07T20:04:06.487Z · LW(p) · GW(p)
Are "white middle class norms" substantially different from, um, black middle class norms, hispanic middle class norms, asian middle class norms and the like?

I'm gonna try to answer this neutrally, but am worried about the risk of touching a political live wire in a context where it's likely to do little good.

The history of the British Empire and then America in the 20th Century suggests that the "middle class norms" being referred to come initially from a particular culture which was mainly made up of white people, and other groups (both residents of other countries and immigrants to Anglo countries) have assimilated these norms in order to participate in the global economic institutions designed by and for people of that culture. It's not obvious that "middle class" as a concept is a cultural universal, much less that middle class norms are the same across cultures.

This doesn't necessarily make such norms objectionable, though for reasons I hope are obvious it's understandable that people are touchy about it all around.

Replies from: totallybogus
comment by totallybogus · 2018-09-07T20:31:19.338Z · LW(p) · GW(p)

...It’s not obvious that “middle class” as a concept is a cultural universal, much less that middle class norms are the same across cultures.

The concept of "middle class" (in the "middle class norms" sense) is increasingly co-evolving with existing cultures in a way that makes it more of a cultural universal. And cultures which don't adopt the middle class concept tend to fail at basic human flourishing, which is as close to a universal as it gets. Marx was well aware of this BTW; he thought socialism would be infeasible unless and until the "middle class norms"-based stage of history (originating from early-modern-age Europe at the latest, not the 20th-century Anglosphere) had fully played out in most of the world, at which point it would be superseded in a quite natural way. See also Scott's post "How the West was won", which is relevant to this question.

Replies from: Benquo
comment by Benquo · 2018-09-07T20:51:33.534Z · LW(p) · GW(p)

I agree that if you assume there's a natural direction of cultural progress, a set of stages all cultures have to pass through (or leapfrog by assimilating into a more advanced culture), then the "middle class" could be a cultural universal in the sense of being an essential attribute of one of those stages. But it's not at all obvious to me that Marx was right. Even if he was, it's not obvious that the actually existing acculturation people do to participate in the global cultural middle class is entirely composed of culturally universal middle-class traits, rather than accidental traits attributable to the particular areas where this culture emerged first.

Replies from: totallybogus
comment by totallybogus · 2018-09-07T21:07:37.009Z · LW(p) · GW(p)

Even if he was, it’s not obvious that the actually existing acculturation people do to participate in the global cultural middle class is entirely composed of culturally universal middle-class traits, rather than accidental traits attributable to the particular areas where this culture emerged first.

Some such traits undoubtedly exist; for instance, people throughout the world learn English for no other reason than to take part in a successful culture where "middle class" traits are relatively common. But it's not clear that there could be any alternative to English that would not be "attributable to [some] particular area"; for example, Esperanto is culturally European and perhaps even specifically Eastern-European; Lojban was indeed designed to be culturally and areally neutral but this doesn't seem to help its popularity, since the Lojban-speaking community is in fact quite tiny.

Replies from: Benquo
comment by Benquo · 2018-09-07T21:27:00.528Z · LW(p) · GW(p)

I agree. Do you think there's some other opinion I should hold differently because of this, either stated or implied?

Replies from: totallybogus
comment by totallybogus · 2018-09-07T21:34:38.658Z · LW(p) · GW(p)

Not necessarily; if anything, I was in fact agreeing with you that some portion of people's 'existing acculturation' to middle-class culture is not, strictly speaking, neutral, due to historical path dependence if nothing else. But I still think it may be unproductive and even pointless for people to act overly "touchy" about such subjects. Should, e.g. Quebeckers, and perhaps Francophones in general, feel justified about their "touchy" attitude wrt. the cultural dominance of English?