Self-Binding

What it is

Physical self-binding means barriers of space: getting the thing out of the house, out of the room, out of reach. Chronological self-binding means barriers of time: limiting consumption to fixed windows, days or milestones. Categorical self-binding means barriers of kind: ruling out a whole class of thing rather than rationing it, on the reasoning that sub-classifications are what a craving negotiates with.

The framing that carries the concept is not the taxonomy but the prerequisite. Lembke insists that self-binding is not primarily a matter of will. The precondition is acknowledging the coming loss of voluntariness while you still have voluntary choice, because in the throes of desire there is no deciding. The pact is made by the version of a person who still has a vote.

It sits alongside three neighbouring ideas this publication keeps separate on purpose. The Ulysses pact is the historical device and the ethical frame. Precommitment is the general behavioural-economics category. Self-binding is the clinical version, cut by space, time and kind rather than by cost, effort and identity, and framed around impending loss of voluntariness rather than around plain temptation.

In effect

The honest part of the treatment is that Lembke then shows the technique failing, twice, at length. She writes a caveat against her own device, noting that self-binding is not proof against self-deception, bad faith or faulty science, and then illustrates the cases where it did not work rather than the successes, at a length that works against her own thesis. That is unusual and is recorded in her favour, and it means this entry should never be used to sell self-binding as reliable.

The attribution is the other thing worth recording, because it is model practice. In her endnote she names her source as Sally Satel and Lilienfeld's 2014 paper in Frontiers in Psychiatry, states that she admires Satel's work, states plainly that she disagrees with the article's premise, and then hands the reader an earlier claimant she did not take the idea from: Thomas Schelling, who described the same manoeuvre as self-management and self-command in the American Economic Review in 1984. A writer naming her source, declaring that she likes its author, declaring that she rejects its argument, and then supplying a prior claimant is doing something almost no popular science author does.

The disagreement underneath is real and is about the definition of addiction rather than about the data. Satel and Lilienfeld read self-binding as evidence that addiction is a matter of choice, since people who bind themselves are exercising agency. Lembke reads the same behaviour as compatible with disease, since the binding is an admission that agency is about to fail. This publication takes Lembke's side, on the grounds that arranging the world in advance precisely because you expect to lose the ability to choose is better evidence for loss of voluntariness than against it, while recording that the technique's existence settles neither position.

A non-clinical instance of the chronological mode comes from military practice, where a fuel state fixed in advance and enforced by an alarm, and a cut-off line on a planning window, do the same work. What the institutional version adds is a witness. Lembke's cases are individuals binding themselves in private; these are individuals bound by an organisation that will find out, so the moment of undoing is observable and gets examined afterwards. A device is only as good as the difficulty of undoing it, and a witness is one of the things that makes undoing harder. The pilot in that account unties himself anyway, which is where the analogy stops flattering the device.

What it does not say

It does not say that self-binding works. No effect size is attached to it here, because the source supplies none, and the source itself illustrates the technique failing.

It does not resolve whether addiction is a disease or a choice. Both readings of the same behaviour are on the record, and the existence of the technique settles neither.

It does not contain instructions. No protocol is described, and anyone considering a change to their own care needs a clinician rather than an entry.

It does not need the depletable-resource theory of willpower to be true. The whole point of binding in advance is that it asks self-control to do nothing at the moment it would fail.

---