Overview
Cumulative Record is a collection of B. F. Skinner’s own papers, essays, addresses, and short technical reports, assembled and arranged by Skinner himself to bring scattered writings together between two covers. It is not a systematic treatise or a textbook — Skinner had already written books of that kind, notably The Behavior of Organisms and Science and Human Behavior — but rather a selection drawn from journal articles, conference papers, and pieces written for broader audiences, intended to show the range of a single research program as it developed. The collection has appeared in more than one edition of differing scope, so its precise table of contents varies; what remains constant is the character of the whole: a working scientist’s sampling of his own output, running from laboratory methodology and the theory of reflex and operant behavior through the technology of teaching and the interpretation of culture.
The title and the organizing idea
The title is not decorative. The “cumulative record” is the sheet of paper produced by the cumulative recorder, the instrument Skinner devised in which a pen steps forward with each response and the resulting curve’s slope gives the rate of responding at every moment. That curve, continuous and always changing, was for Skinner the model datum of a science of behavior: it displays behavior as it actually occurs rather than as an average computed after the fact. By borrowing the phrase for the book, Skinner presents the volume as a comparable record — the accumulated trace of one investigator’s own behavior over time, piece by piece, in the order in which the work accumulated.
The experimental analysis of behavior
The core of the volume consists of papers that establish and defend the fundamental concepts of the experimental analysis of behavior. “The Concept of the Reflex in the Description of Behavior” is a historical and methodological essay that treats the reflex not as an inferred nervous event but as a lawful correlation between stimulus and response, defined operationally and stripped of hypothetical physiology. “The Generic Nature of the Concepts of Stimulus and Response” argues that stimulus and response are classes of events united by functional relations rather than by physical similarity — a point of central importance, since the whole program depends on treating many different lever presses or key pecks as instances of one response. “Two Types of Conditioned Reflex and a Pseudo-type” sets out the distinction between respondent (Pavlovian) conditioning, in which a stimulus elicits a response, and operant conditioning, in which a response is emitted and controlled by its consequences.
Alongside these conceptual papers sit empirical ones, most famously “‘Superstition’ in the Pigeon,” the study in which food delivered at fixed intervals regardless of what the bird was doing produced ritualistic, idiosyncratic movements that persisted as though they had produced the food. The paper is both a demonstration of accidental reinforcement and a compact illustration of Skinner’s preference for demonstrating a principle in a controlled experiment rather than arguing for it discursively. Other papers in the volume treat rate of responding as the fundamental measure, the analysis of schedules of reinforcement, and the extension of these methods to the study of psychotic patients and to the assessment of drug effects — the beginnings of behavioral pharmacology.
Method and theory
Two essays in the collection are among the most widely cited of Skinner’s programmatic statements. “Are Theories of Learning Necessary?” attacks the appeal to hypothetical internal processes, intervening variables, and neurological “explanatory fictions,” arguing that the proper business of a science of behavior is the description of functional relations between environmental events and behavior, and that premature theorizing distorts both research and interpretation. “A Case History in Scientific Method” is a reflective, partly autobiographical account of how Skinner’s research program actually proceeded — the accidents, the improvised apparatus, the decisions to follow a surprising result rather than a plan. It contains his much-quoted advice to the researcher who runs into something interesting to drop everything else and study it. Read together with “The Operational Analysis of Psychological Terms,” with its treatment of private events and verbal report, these papers give the collection its methodological spine.
Application: teaching, language, and the design of culture
A substantial portion of the book is addressed to what Skinner called the technology of behavior — the deliberate use of behavioral principles outside the laboratory. “The Science of Learning and the Art of Teaching” and “Teaching Machines” are the classic statements of programmed instruction: the analysis of classroom learning in terms of small steps, active responding, immediate feedback, and self-pacing, together with the mechanical devices designed to embody those requirements. Other pieces extend the same reasoning to clinical settings, to the interpretation of psychotic behavior as a product of aversive control and faulty contingencies, and to psychoanalytic concepts, which Skinner treats as redescriptions that obscure the environmental variables doing the real work.
The volume also includes the essays for which Skinner was most publicly known and criticised. “The Control of Human Behavior” and “Freedom and the Control of Men” argue that behavioral technology is already in use, that the choice is not between control and no control but between effective and haphazard control, and that a scientific culture should design its practices deliberately. Related papers on the design of cultures and experimental communities extend the argument from the individual to the social group. These pieces are polemical in tone and are among the most accessible in the book.
Verbal behavior and literary analysis
Skinner’s interest in language is represented not only by technical papers on verbal behavior but by two unusual quantitative studies of poetry: an analysis of alliteration in Shakespeare’s sonnets and a quantitative estimate of sound patterning in verse. In these, Skinner counts and classifies the sound features of literary texts in order to explain the effects of poetry in behavioral terms, treating the literary object as a set of stimulus variables acting on a reader. They are idiosyncratic pieces — a psychologist’s foray into criticism — but they are characteristic of the collection’s insistence that the same principles apply to the highest cultural products as to a pigeon’s key peck.
Character of the collection
Because the pieces were written at different times, for different audiences, and with different degrees of technicality, Cumulative Record reads less as an argument than as a record of a career’s activity. The book’s value to readers lies precisely in that heterogeneity: it makes available in one volume papers otherwise buried in journals, and it shows the same set of assumptions being applied to apparatus design, to the concept of the reflex, to classroom instruction, and to the question of how a society should organize itself. It is a book of essays and articles by a single author, unified by a method rather than by a narrative, and it is best approached as a sourcebook of Skinner’s own statements of his position rather than as an introduction to it.
Reinforcement as selection
The technical papers give the collection its experimental center. Reinforcement is treated not as a reward added to behavior but as a selective operation: responses followed by certain consequences become more frequent, and the resulting patterns depend on the schedule and context in which reinforcement occurs. The book documents a shift from studying isolated responses to studying ongoing streams of behavior, where extinction, discrimination, generalization, and intermittent reinforcement are aspects of a single contingency analysis rather than separate topics. The laboratory apparatus and recording methods are presented as ways of making that stream visible, so that behavior can be described in terms of rate, probability, and change over time rather than in terms of hidden faculties.
Against explanatory fictions
A recurring target is the tendency to explain behavior by positing internal agents — learning, memory, drive, purpose, the self — and then treating those terms as causes. The essays argue for a functional analysis: identify the environmental variables of which behavior is a function, and state the relations between behavior and consequences. This is not a denial of private events. It is an attempt to ask how such events can be talked about within a behavioral vocabulary and how mentalistic terms can be reinterpreted as labels for repertoires under stimulus control. The same critical stance is applied to theories of learning, where the need for explanations that go beyond the description of functional relations is questioned.
Interpretation and the extension of principles
The collection distinguishes between experimental analysis and interpretation. Where variables cannot be manipulated directly — in complex human conduct, in language, in social arrangements — the strategy is to extend principles established in the laboratory and to see how far they account for the phenomenon. This is why essays on instruction, clinical interpretation, and cultural design sit alongside animal experiments without embarrassment: they are exercises in interpretation, not substitutes for experimental proof. The book’s coherence is rhetorical as much as doctrinal; the same assumptions are restated for different audiences and different problems.
The title and the record
The title carries a double sense that illuminates the whole. A cumulative record is the graphical trace produced by operant apparatus, showing behavior as a continuous quantity changing under contingencies; the volume is also a cumulative record of one behavioral scientist’s public statements. Reading it in that light explains its range: technical reports, methodological polemics, popular essays, and speculative extensions all belong to the same record. Its significance is less as a systematic textbook than as a documentary trace of behaviorism’s ambitions — to make behavior the subject matter, to replace explanatory fictions with contingencies, and to treat the laboratory as a base from which to interpret culture.
