Template:M intro design thin rules
In her illuminating book Rules: A Short History of What We Live By Lorraine Daston distinguishes between “thick rules” and “thin rules” as means of understanding the regularities of daily life.
Thick rules are principles; heuristics; general statements that require interpretation and construal: judgment in light of prevailing circumstances. End-users must decide for themselves how to act. “Cross a road when it is safe to do so”. Rules of this kind depend on the intelligence, experience and trustworthiness of those they govern (both drivers and pedestrians): users must have their own model of when, and what counts as “safe”.
A free market is a system of thick rules. It depends on participants’ trust and good faith. “Credit” means, literally, “she believes”. There is no mandated conduct in a given circumstance, so markets imply a degree of interpersonal vulnerability, open as they are to abuse by the unscrupulous. Trustworthiness — good eggness — is a critical value in thick-ruled communities: without it the thick rules, and quickly thereafter the community itself, will deteriorate.[1] This interpersonal vulnerability is a feature, and not a bug, in the system: there is a good argument that society itself depends on — or, really, is — a mutual framework of undischarged interpersonal debts.[2] In any case, we have a series of soft-edged principles with leaky boundaries, and that we forebear from taking advantage of those weaknesses is what makes our polity strong.
Governments come and go in how far they trust their constituents to do the right thing. When it seemed like we were near the end of history, Reagan campaigned that “the Government is the problem”. The series of geopolitical events — 9/11, the global financial crisis, COVID, the Russia/Ukraine conflict has persuaded regulators that society cannot be trusted to look after itself, and meanwhile the information revolution has vouchsafed tools and techniques increasingly permitting granular regulation, and we find ourselves in anno domini 2023 at the far end of the pendulum’s arc. Thick rules are in decline; wherever we look there are thin rules minutely prescribing behaviour and micromanaging risk.
Thin rules are algorithms and not heuristics. They are specific and not general. They are deterministic and not aspirational. They take out doubt, remove uncertainty, and leave nothing to trust, judgment and interpersonal vulnerability.
Knowledge and skills
Interviewed on Matthew Taylor’s Forward Thinking podcast, Daston discusses the relationship of thick rules and thin rules to knowledge and skills, and it occurs that this discussion could be transplanted quite easily onto the agency problem, particularly as it relates to the practice of law.
In essence thick rules require skills to apply them; thin rules demand only the knowledge required to obey them. The former invites some kind of dynamic interaction between human and environment; the latter is more a case of symbol processing.
This is precisely what an expert agent wants: a customer to whom it can supply rules comma and which the customer need only the wear with all to be able to competently follow, and who is therefore obliged to return to the agent repeatedly for refreshers about different circumstances. The agency model is to give — well, sell — a man a fish rather offering to teach him to fish.
But at least when it comes to business as usual activity, this kind of specialisation, in Robert Heinlein’s phrase, is for insects. We should aspire to ensure our subject matter experts have skill and not just knowledge when it comes to carrying out business as usual. Buy the fish, that is to say, only when a special banquet calls for caviar.
The daycare experiment as an illustration of thin rules
Uri Gneezy’s famous daycare experiment points up the difference nicely.[3]
A daycare centre, frustrated that parents were arriving late to pick up their children, imposed a fine on those arriving more than ten minutes late. Far from eradicating lateness, parental delinquency got worse, not better. Parents now felt there was a price to their lateness, which they were prepared on occasion to pay.
The daycare unwittingly converted a thick rule into a thin one. The fine was a proxy for the essential moral principle designed to achieve an equivalent outcome in a measurable way, in the form of a penalty. It was nicely clear and measurable but, being a simple trigger, was still a poor derivative of the complex sociological circumstances from which the underlying moral principle was constructed.
But the amount of the penalty and the point at which it applied were arbitrary. Neither had anything to do with the welfare of daycare centre staff: what difference was there whether parents were 14 (free) or 16 (charged) minutes late? What difference is there whether the parent has one child or two? Is the penalty even passed to affected staff? If it is, does it fairly compensate each for her own specific, personal inconvenience?
At the same time the thin rule imposed obligations upon parents, it also conferred clear rights upon them. Before it was introduced, the parents had an unconditional obligation to pick their children up on time, and no concomitant right to expect staff to wait for them should they be late. Staff would, of course, wait — crises happen, after all — and even though not not obliged to compensate the staff, parents would still incur some “social indebtedness” as a result which, in a strong community, would be subsequently reciprocated (a thank-you gift, a lift home, a return favour; some opportunity to help out later, and so on) and these undischarged informal mutual debts, for which there is no detailed accounting, themselves strengthen the community.[4]
But as soon as it as a payable sum, there are now equivalent, equally arbitrary rights which parents may seek to enforce. I have paid my 10 shekels, now you must stay for an hour: this is my entitlement. The arbitrary boundaries drawn by the thin rule become a quantifiable value, irrespective of the substance of the circumstances. We have created expectations where there are none.
Compliance culture
The same goes for our modern compliance culture. We replace “principles-based regulation” with detailed rules, hire a raft of outsourced drones to implement those regulations, and create careers for people whose job it is to enforce, analysing and defend and ensure compliance against formal rules. These compliance and audit staff need no grasp of the fundamental moral principles the rules they police are designed to deliver: they are complex sociological circumstances which require judgment and experience we should not expect compliance officers to have. and in any case the fundamental moral principals are beside the point. They are substituted by granular rules. The granular rules are all that matter: it is a distraction to understand the underlying principle. Where the two conflict it is a disadvantage. We see it again: the generational refrain: substance succumbs to form.
- ↑ See Adam Curtis’ terrifying Traumazone for a vivid illustration of a society — Russia — that has lost all trust in its participants and institutions.
- ↑ See David Graeber’s magnificent Debt: The First 5,000 Years.
- ↑ A Fine is a Price, Uri Gneezy & Aldo Rustichini, Journal of Legal Studies, Vol. 29, No. 1 (January 2000). Mentioned also in Dan Ariely’s Predictably Irrational: The Hidden Forces that Shape Our Decisions.
- ↑ Graeber is very good on this point too: the last thing a community wants is to finally settle all mutual indebtedness as it is the very thing that binds the community together. There is an argument that a community is little more than a group of individuals with inarticulable mutual debts.