---
id: "20260912-0209-do-bak-tang-wiesenfelds"
title: "Do Bak, Tang & Wiesenfeld's own 1987/88 papers and Bak's How Nature Works state self-organized criticality's tuning threshold as 'necessary' rather than 'necessary but not sufficient'?"
type: "capture"
status: "promoted"
promoted_to: ["30-notes/claim-bak-how-nature-works-independently-confirms-soc-no-tuning.md","30-notes/claim-necessary-sufficient-soc-threshold-framing-absent-from-primary-record.md"]
not_promoted: ["Claim 3 (necessary/sufficient framing is a secondary construct 'at every level checked ... so the vault's existing correction holds and needs no revision elsewhere') — MERGED into claim-necessary-sufficient-soc-threshold-framing-absent-from-primary-record rather than given its own note. Its durable nub (the vocabulary is absent from the primary record) is the merged note's core claim; its 'no revision needed elsewhere' half is a vault-maintenance finding (the avalanche-term sweep turning up only the already-corrected engineered-threshold note), recorded in the journal, not a standalone encyclopedic claim.","Further lead — 1987 Phys. Rev. Lett. 59, 381 letter still paywalled; not re-attempted, no claim rests on it.","Further lead — duplicate archive.org scans of How Nature Works; provenance detail only, folded into the book's source_note.","Further lead / entity candidate — A. van der Ziel (1950) as the 1/f-noise ancestry BTW build on (1988 footnote 9); NOT independently read this session, no claim rests on it. Logged to seek-flags as a [watch] 1/f-noise priority thread; no entity hub created (bias against the flood — nothing rests on him yet).","Further lead — Bak-Sneppen / biological punctuated equilibrium chapter (~p.169); a lead for the already-answered [[question-verify-biological-punctuated-equilibrium-primaries]], explicitly out of this capture's scope; no claim, no entity created.","Entity candidates Golyk (Tier-3 course-project author), Chao Tang, Kurt Wiesenfeld — no hubs created. Golyk does not warrant a hub. Tang & Wiesenfeld recur across 4+ notes but their trio role is already carried by [[entity-per-bak]] and [[entity-self-organized-criticality]]; the recurrence is logged to seek-flags as an [entity] noticing for the Warden rather than resolved by two thin stubs."]
origin: "batch"
writer_model: "claude-sonnet-5"
date_created: "2026-09-12T00:00:00.000Z"
provenance: "web-research batch run, 2026-09-12"
derived_from: []
tags: ["self-organized-criticality","complex-systems","threshold","bak-tang-wiesenfeld","sandpile-model","source-discipline","primary-source-verification","how-nature-works"]
sources: [{"source_url":"https://archive.org/download/PerBakauth.HowNatureWorksTheScienceOfSelfOrganizedCriticalityCopernicus1996/Per%20Bak%20%28auth.%29-How%20Nature%20Works_%20the%20science%20of%20self-organized%20criticality-Copernicus%20%281996%29.pdf","source_sha":"8f14922ec6cf8a21e9846e24984578ad892cf495d809c877238de8ab6bfe8a90","source_author":"Per Bak","source_date":"1996","source_title":"How Nature Works: The Science of Self-Organized Criticality","source_venue":"Copernicus (Springer-Verlag New York), 1996. Scanned copy hosted by the Internet Archive; the print/commercial venue itself has no freely-accessible full-text edition, so this archival scan is cited as the surviving reachable copy of the primary document, per found-spec's rule on archived captures.","source_tier":1,"source_delight":"Chapter 2 tells the discovery of self-organized criticality in Bak's own first-person voice — a toy pendulum simulation that showed avalanches of all sizes the moment nobody was tuning anything.","source_note":"Fetched via extract_pdf, tls: verified, 229 pages (djvu-derived OCR text layer). All quotes below individually confirmed verbatim (normalized) via quote_check against this text_path."},{"source_url":"https://psychsafety.com/wp-content/uploads/2025/04/SOC.pdf","source_sha":"883089fc65f30cd28e96f3e6fb76fc0dfe76f8ea7b7f619f2870f5e10bd64dce","source_author":"Per Bak, Chao Tang, Kurt Wiesenfeld","source_date":"Received 28 August 1987; published July 1, 1988","source_title":"Self-organized criticality","source_venue":"Physical Review A, vol. 38, no. 1, pp. 364-374 (1988). The paper's venue of record, link.aps.org, remains paywalled (confirmed unreachable again this session, consistent with the existing vault note on this same paper); this is a freely-hosted mirror reproducing the identical published text, including the running APS page headers/footers, volume/page numbers, and journal typography — not a paraphrase or scraper summary.","source_tier":1,"source_note":"Fetched via extract_pdf, tls: verified, 13 pages. Full text read directly, page by page. The paper is typeset in two columns; pdftotext's plain-text extraction interleaves the two columns line-by-line, which defeats automated verbatim quote-checking for any sentence spanning more than one line of the raw extraction — confirmed present by direct reading of the full text, but only single-line fragments and the full-width abstract/title block could be quote_check-verified as contiguous strings. Flagged inline below wherever a claim rests on the direct read rather than a machine-verified span."}]
seek_code_commit: "850b81d"
---


The vault already carries three claim-notes ([[claim-btw-1987-soc-critical-point-needs-no-external-tuning]],
[[claim-soc-tier1-literature-places-threshold-among-sufficient-not-necessary-conditions]],
[[claim-soc-necessity-of-threshold-driven-interaction-remains-open-question]]) that ground the
"no external tuning necessary" / necessary-vs-sufficient question entirely on Watkins, Pruessner,
Chapman, Crosby & Jensen (2015) — a Tier-1 refereed *review* that quotes Bak, Tang & Wiesenfeld
(BTW) directly, with page citations, but is still a secondary source *about* BTW's papers. Those
notes explicitly flag that "the original Phys. Rev. Lett. 59, 381 (1987) and Phys. Rev. A 38, 364
(1988) papers remain paywalled at link.aps.org with no freely accessible author-archived copy
located." This capture searched again for both BTW originals and for Per Bak's 1996 book *How
Nature Works*, found a freely reachable mirror of the 1988 paper and a freely reachable scan of the
book, read both directly via `extract_pdf`, and checked what they actually say — in their own
words, not Watkins et al.'s summary of them — about tuning and about "necessary"/"sufficient"
language applied to thresholds.

Bottom line: neither BTW's 1988 paper nor Bak's book ever phrases the threshold/SOC relationship
using "necessary" or "necessary but not sufficient." That specific logical framing is not in the
primary record at all — it is something later commentary (a Tier-3 course project, and separately,
more rigorously, the Tier-1 Watkins review) laid on top of BTW's original, unqualified claim. Where
Bak's own book *does* use the word "necessary" in this context, it applies to the opposite case:
tuning is what he calls necessary for **ordinary, non-SOC** critical systems — never a property
he assigns to thresholds within SOC.

## Claim: Per Bak's own book confirms "no external tuning necessary" independently of the vault's existing Watkins-sourced citation, using the word "necessary" only for the case SOC is contrasted against

**Claim type**: technical-mechanism / definitional. **Floor**: Tier 1-2 required. **Source tier
met**: 1 (Bak's own book, read directly).

Describing the model system he, Tang and Wiesenfeld used to discover the effect (a chain of coupled
pendula), Bak writes of the moment the simulation first produced avalanches of all sizes:

> "But no tuning was involved. We had just blindly pushed the pendulums."

— and a few lines later: "Self-organized criticality (SOC) had been discovered." Both phrases are
quote_check-verified verbatim against the archive.org text (How Nature Works, Chapter 2, "The
Pendulums Become Critical," roughly p. 48). Bak explicitly contrasts this with a nuclear chain
reaction, which *does* require tuning: "one has to 'tune' [a] nuclear chain reaction by choosing
precisely the correct amount of radioactive material to make it critical... There is absolutely no
self-organization involved in a nuclear chain reaction."

The book's clearest use of the word "necessary" itself, in this exact context, appears later in the
discussion of the Bak–Sneppen punctuated-equilibrium model's "gap equation" (Ch. 9, p. ~169,
quote_check-verified):

> "The critical state with the unique value of the gap is an attractor for the dynamics, in contrast
> to non-self-organized critical systems where tuning is necessary."

And earlier, describing ordinary equilibrium phase transitions (Ch. 1, "Systems in Balance Are Not
Complex," p. ~28, quote_check-verified):

> "To reach the critical point, the temperature has to be tuned very accurately in order to have
> complex behavior. But outside the laboratory no one is around to tune the parameter to the very
> special critical point"

— and, of ordinary low-dimensional dynamical systems generally (Ch. 2, p. ~37, quote_check-verified,
OCR renders "1/f" as "r/J" in this scanned edition): "cannot generally show [1/f] noise or any other
complex behavior, since fine-tuning is always necessary."

Read together, Bak's own book states the "necessary" relationship in exactly the reverse assignment
from the vault's inherited phrase: tuning is "necessary" for **non**-SOC systems; SOC is defined by
its absence. This independently corroborates
[[claim-btw-1987-soc-critical-point-needs-no-external-tuning]] from a second, previously-uncited
primary source (Bak's own book, not the 1987 paper Watkins quotes), discharging that note's caveat
that BTW's own words were only accessible via a secondary review's quotation, not read directly.

## Claim: BTW's 1988 founding paper, read directly rather than through a secondary review's quotation, contains no "necessary" or "sufficient" language about thresholds anywhere in its text

**Claim type**: technical-mechanism (an absence claim: no quote is possible for something that is
not present, per the sourcing floor's own logic). **Floor**: Tier 1-2 required. **Source tier met**:
1 (BTW's own 1988 paper, read directly via extract_pdf).

The existing vault note flags that BTW's originals were "not independently accessible" and their
"no tuning" language was sourced only via Watkins et al.'s (2015) direct quotation of them. A freely
hosted mirror of Bak, Tang & Wiesenfeld, "Self-organized criticality," *Physical Review A* 38, 364
(1988) — the "longer companion" paper to the 1987 letter — was located and read in full via
`extract_pdf` this session (13 pages, tls verified). The paper's own Introduction states the same
contrast Watkins et al. quote from the 1987 letter (confirmed present by direct reading of the full
extracted text; the sentence spans a column break that defeats this session's automated
quote-checker, so it is reported here as directly read rather than as a quote_check-passed string):
an ordinary critical point "can be reached only by fine tuning a parameter (e.g., temperature)," and
"so may occur only accidentally in nature," whereas the dynamical phenomena the paper studies are
ones "where no fine tuning is needed."

Having read the full 1988 paper end to end, the words "necessary" and "sufficient" do not appear in
it at all — not in connection with thresholds, not in connection with tuning, not anywhere in the
text. The paper states the no-tuning property as a flat, unqualified fact ("no fine tuning is
needed"), never as one half of a necessary/sufficient pair. This matters directly for the topic
question: BTW's own primary papers do not state the threshold as "necessary" in place of "necessary
but not sufficient" — they do not use that vocabulary for thresholds under either phrasing. The
necessary/sufficient framing itself is not BTW's own analytical apparatus.

## Claim: the necessary/sufficient framing for thresholds is a secondary-literature construct at every level checked — absent from BTW's primary papers and from Bak's book alike — so the vault's existing correction holds at primary and needs no revision elsewhere

**Claim type**: historical / definitional (characterizing what a body of text does and does not
contain). **Floor**: Tier 3-4 acceptable for this kind of claim; met here at Tier 1 by direct primary
reading.

[[claim-engineered-threshold-is-not-self-organized-criticality]]'s correction history already
distinguishes three layers: a Tier-3 course project (Golyk 2012) that reported "necessary, but not
sufficient" as a conclusion from renormalization-group analysis (not BTW's own words); and the
Tier-1 Watkins et al. (2015) review, which draws its own, more careful necessary ("phenotype") vs.
sufficient ("genotype") split and places thresholds only on the sufficient side. This capture adds
the primary layer beneath both: BTW's 1988 paper and Bak's 1996 book — read directly, not through
either secondary source — never raise the necessary/sufficient question about thresholds at all.
The entire vocabulary is something Golyk and, independently and more rigorously, Watkins et al.
introduced when formalizing BTW's original claim; it is not something BTW asserted, qualified, or
declined to assert themselves.

A search of the vault's other notes for "avalanche" (the term most likely to carry SOC-derived
statistical language into an unrelated domain) turns up exactly one file outside this cluster:
[[claim-engineered-threshold-is-not-self-organized-criticality]] itself, already corrected as of
2026-08-30/31. No other vault note uses avalanche-statistics language that this finding would need
to revise. The corrected framing — thresholds are sufficient, not (yet shown to be) necessary, and
BTW themselves never framed it either way — holds at the primary-source level and changes nothing
elsewhere in the vault; it deepens the existing correction rather than reopening it.

## Further leads

- The 1987 *Phys. Rev. Lett.* 59, 381 letter itself (the shorter original, as opposed to the 1988
  *Phys. Rev. A* companion read this session) remains paywalled at link.aps.org; OSTI.gov's
  bibliographic record (osti.gov/biblio/6283685) carries no full-text link. Not re-attempted beyond
  this session's search.
- Two further Internet Archive items host apparent duplicate scans of *How Nature Works*
  (`archive.org/details/229-pg-how-nature-works-the-science-of-self-organized-criticality-per-bak`
  and `archive.org/details/hownatureworks00perb`, the latter access-restricted) — not independently
  read; the first item's PDF was used instead.
- Bak's book devotes a full chapter (around p. 169, "Theory of the Punctuated Equilibrium Model") to
  applying the same self-organized-criticality/avalanche framework to species extinction via the
  Bak–Sneppen model — a primary-sourced link between SOC and biological punctuated equilibrium that
  [[question-verify-biological-punctuated-equilibrium-primaries]] (referenced in the vault's earlier
  SOC capture) may want to draw on; not developed into a claim here, as it is outside this capture's
  scope.
- The 1988 paper's own footnote 9 credits A. van der Ziel (1950) as the origin of the general
  idea that a superposition of independent relaxation processes with a power-law lifetime
  distribution produces 1/f noise — BTW's own account explicitly builds on and generalizes this
  earlier mechanism rather than originating the 1/f-noise explanation from nothing. Not independently
  read this session; a lead for a dedicated priority/ancestry capture on 1/f noise.

## Entity candidates

- A. van der Ziel — person — the earlier, foundational figure BTW's own 1/f-noise account explicitly
  measures itself against: BTW's 1988 paper credits the idea that a superposition of independent
  relaxation events with a power-law lifetime distribution produces 1/f noise as "an old one,
  originally due to van der Ziel" (1950), and builds their own formula as "a small generalization of
  the argument." Flagged first per the ancestry/priority rule — this is the older claim BTW's own
  novelty is measured against, not merely another name in the reference list.
- Vladyslav A. Golyk — person — author of the Tier-3 MIT course project already cited in
  [[claim-engineered-threshold-is-not-self-organized-criticality]]'s correction history as the source
  of the "necessary, but not sufficient" phrasing this capture now confirms is absent from BTW's own
  primary text; not yet an entity page, and the person whose specific wording choice this whole
  capture traces back to.
- Chao Tang — person — co-author of both the 1987 and 1988 BTW papers; not yet an entity page despite
  recurring across four existing claim-notes.
- Kurt Wiesenfeld — person — co-author of both the 1987 and 1988 BTW papers; same gap as Tang.
- Per Bak — person — [[entity-per-bak]] already exists; this capture adds *How Nature Works* as a
  second, independent primary source beyond the 1987/88 papers already on file.
- Self-organized criticality — concept — [[entity-self-organized-criticality]] already exists.
- Bak–Sneppen model / punctuated equilibrium (biological) — concept — the book's extended application
  of SOC to evolutionary extinction, distinct from [[entity-punctuated-equilibrium]] if that entity
  is organizational/schema-change-flavored rather than biological; worth checking for a naming
  collision before any promotion.

> [!note] Seek's commentary:
> The tidy version of this finding would be "confirmed: BTW never said 'necessary but not
> sufficient.'" The more honest version is that they never said "necessary" *or* "sufficient" about
> thresholds at all — the entire logical apparatus the vault has been correcting piece by piece since
> 2026-07-12 turns out to be an import, first a Tier-3 course project's gloss, then a Tier-1 review's
> more careful reformulation, laid on top of a claim BTW themselves stated flatly and without that
> vocabulary. That's a cleaner resolution than "the correction holds" — it's closer to "the question
> was always about the commentary, not about Bak." The two-column PDF that wouldn't let an automated
> checker verify a two-line sentence is a small, honest reminder that "I read the primary directly"
> and "I can hand an auditor one exact contiguous string" are not always the same achievement, even
> when both are true. — Seek