Use "none" when "some" is harder
A clean zero is often easier to sustain than a tightly rationed "a little."
Why it works
Allowing "some" means re-deciding the amount every single time, and each decision both depletes effort and invites escalation. A categorical "none" removes the quantity question entirely — there is no "how much" to negotiate, only "do I cross the line." For genuinely tempting things, abstinence can paradoxically demand less self-control than moderation.
How to do it
- For temptations where one leads to more, set the line at zero rather than a small allowance.
- Notice when "just a little" reliably becomes "a lot" — that is the signal to choose none.
- Reserve moderation for things you can genuinely stop at a set amount.
Evidence
Consistent with clinical observation in addiction treatment that abstinence can be easier than controlled use for some people, and with the general finding that fewer decisions reduce failure. The "none beats some" claim is person- and substance-dependent, not universal. (clinical)
For some people and behaviors, moderation is both possible and healthier than rigid abstinence; "none" is a tool, not a moral rule.
Common mistake
Applying "none" reflexively to everything, including things you could moderate fine, which creates needless deprivation and rebound.
Practice this with IX Coach
7 days free, then $40/month (~$1.30/day).
More practices for Bright-Line Rules: When "None" Beats "Some"
- Draw one clear, binary line
Replace "less" or "moderate" with a rule that is unambiguously on or off.
- Close the loophole-by-loophole negotiation
Pre-decide so the moment of temptation has nothing left to argue about.
- Make the line and its crossings visible
A bright line only works if you can clearly tell when it is crossed — so track it.
- Define exceptions in advance, in writing
A bright line with a pre-written exception stays bright; an improvised exception breaks it.
- Know when a bright line backfires
Rigid all-or-nothing rules can fuel disordered patterns — use them with judgment.