Yvonne Barnes-Holmes, Dermot Barnes-Holmes, Louise McHugh and Steven C. Hayes (2004), International Journal of Psychology and Psychological Therapy 4(2):355–375 — no DOI; publisher’s open-access PDF.
Key takeaways
- The theory in brief. Relating one event to another “arbitrarily”, by social convention rather than by the events’ physical properties, is learned operant behaviour. It is trained across many exemplars and comes under the control of cues such as “is”, “more than” or “opposite”. Its products, the relational frames, share three properties: mutual entailment, combinatorial entailment and the transformation of stimulus functions.
- Why it matters for psychopathology. Because human verbal relations are bidirectional, describing a trauma can carry the trauma’s aversive functions. Because functions transform through comparative relations, a man whose panic began in a small local store can rank Macy’s, where he has never been, as more frightening. Neither follows from direct conditioning alone, which is the paper’s argument that behaviour therapy needs RFT.
- What it does to ACT. Acceptance works when a rule that predicts the discomfort coheres with what then happens. Defusion trains a shift from I HERE NOW to I THERE THEN, and a value is the top of a hierarchical network. The paper offers these as heuristics for later work, not as findings.
Standing in the record
Filed on 2026-10-03 at the owner’s request, as the read substitute for the
book that founded Relational Frame Theory. That book is Hayes,
Barnes-Holmes and Roche (2001), LIT-549, filed Deferred beside this
note. The extends relation to it is the paper’s own construction: it
presents itself as describing “the behavioral account of human language and
cognition, known as RFT” (p. 357) as the book set it out, and it cites the
book’s chapters for each concept, so it could not stand without it.
Why this paper and not another statement by the same group:
- Same authors. Dermot Barnes-Holmes and Steven Hayes edited the book and wrote or co-wrote most of its chapters.
- It states the theory. Part 2 covers bidirectional relations, networks, derived relational responding as operant behaviour, the frames and their three properties, relating relations, rules and the self.
- It carries the theory to the batch’s subject. The editors’ reply to their critics (2003, Maynooth deposit) is shorter and is the place to see RFT defended. But it does not say what RFT implies for psychopathology, which is why the batch files RFT at all.
It is the record’s bridge from RFT to the clinical line:
- The 1996 experiential-avoidance paper (LIT-551). This paper cites it for the behaviour analysts’ preference for functional dimensions over syndromal diagnosis (p. 356). It also cites it, as “Hayes, et al. 1996”, where it introduces ACT (p. 369).
- The 2006 ACT review (LIT-555). That review summarizes the same RFT model of psychopathology in one paragraph and refers the reader to the book. This paper is the fuller statement it leaves out.
- Bandura (1977), LIT-545. The paper names Bandura (1977) among the attempts at “integrative models that combined both cognition and behavior” that followed Chomsky’s critique (p. 358). RFT’s own answer to that critique is the rest of the paper.
No instruction for machine-learning practice, and no anthology topic holds a behaviour-analytic theory of language.