A Model of Persuasion

Despite our casual compartmentalization of many diametrically opposed beliefs, humans are also sensitive to cognitive dissonance that is merely mild. The theoretical resolution of this perplexity is that abstract beliefs conflict only insofar as they support opposed concrete opinions. This opens the question of the psychological nature of this support. We can simplify explanation of persuasion if we conceive of the influence of the abstracting mindset on the concretizing mindset within an individual as being the same type of influence as that of an outside persuader on another’s (abstract) beliefs. All direct influence of belief (whether the belief belongs to oneself or another) is suggestion, which in its purest form is hypnosis. The mechanism by which belief influences opinion (primarily as it relates to volition) is autosuggestion. The process by which concrete opinion directly influences other concrete opinions is the ordinary sphere of learning from experience. All this is shown in the diagram below:  

Notice that the diagram shows two routes to belief change: blue, from opinion to belief and using dissonance modulation (as well as ordinary learning); and green, from belief to opinion and using suggestion. Only the green route affords the possibility of rational belief change, and that is only a possibility. This is because logic is approximated only in the concretizing mindset.

The nature of ordinary learning isn’t itself my concern here, just the peculiar relationship between the concretizing and abstracting mindsets. Governing this relationship are two basic processes: suggestion and dissonance modulation. The main practical relevance for writers is that what works for changing belief through suggestion fails for change by dissonance modulation. For example, suggestion demands the greatest simplicity because suggestion involves bypassing the critical faculty (that is, the brain’s cingulate), and the easier a proposition is to understand, the more it is accepted automatically, since acquiring disbelief necessitates express rejection. On the other hand, effective dissonance modulation requires univocality, since without it the recipient of the influence is likely to achieve consonance through a different route than the persuader intends. The difference is that the subject of suggestion submits to influence, whereas the recipient of a communication using dissonance modulation will only accept the writer’s ideas if they are actually dissonance reducing. Imprecision arouses rather than resolves dissonance. As for suggestion, in hypnosis vague suggestions work better than precise ones. (Thus, telling subjects that that they are “going to sleep” can induce hypnosis even though hypnosis isn’t actually much like sleep.)

Conflation of suggestion and dissonance modulation runs deep. It isn’t just a matter of a superficial trend in writing pedagogy; it also afflicts those whose business it is to know better, such as social psychologists and (remarkably) specialists in hypnosis. The bias of social psychologists reflects the orientation to advertising. Although the theories of cognitive dissonance and construal level come from social psychology, the classic social psychological research on persuasion, which provides the framework for general discussions of the persuasion process, implicitly equates persuasion with suggestion. (An equation that is actually worth retaining if its limits are understood, inasmuch as the act of persuading others can be contrasted with the act of convincing them, the blue route contrasting with the green.)

The confusion is greatest within the dominant school of professional hypnotists. The font of clinical hypnosis (outside of hypnoanalysis), Eric Ericson, attributes results he achieves through the artful modulation of cognitive dissonance to some form of hypnotic suggestion. Now, here’s confusion enough to be funny. Dilbert cartoonist Scott Adams, who is a trained hypnotist (apparently of the Ericsonian persuasion), concluded that Donald J. Trump’s methods and results make him a “master persuader.” Perhaps subliminally recognizing that cognitive dissonance is the key to deep rather than superficial attitude change (which is to say, change of opinion rather than belief), Adams took Trump’s ability to use suggestion on supporters, predisposed to accept his influence, to prove Trump had mastered cognitive dissonance. Adams predicted that Trump would win in a landslide because he could hypnotize most anyone.

Lexicographers and usage experts expatiate on the distinction between the verbs persuade and convince. The distinct meanings may be on the verge of loss, but it marks an important psychological difference. According to the lexicographers, persuasion is aimed at obtaining action, as in persuading a judge to sustain a motion. Persuasion means change in belief, without any change in opinion being necessary. But you will rarely change a judge’s belief without changing his opinion. (The effect of suggestion is actually obtained primarily from respective law firms’ prestige.) So, despite the aim being to persuade judges, advocates typically must convince them—of something. For the same reason, when academics try to persuade editors to publish a paper, in the usual they must convince them, at least in the best journals. Even more than when a lawyer influences a judge, the academic must use dissonance modulation, since the practice of blind review screens out many indicia of status that are so influential in the use of suggestion.

This is a general model of the persuasion process (more precisely, of the processes of persuading and convincing) that highlights the relationship between construal-level theory's abstracting and concretizing mindsets. The two epistemic attitudes, belief and opinion, are subject to their respective modes of influence, suggestion from belief to opinion and cognitive-dissonance modulation, from opinion to belief. In the diagram the blue arrows represent the path of cognitive dissonance modulation and the green arrows of suggestion. Hypnosis is a short-cut to belief formation in that it bypasses opinion. Post-hypnotic suggestions are incorporated into the subject’s belief system, whereas the justification is extemporized when an explanation is requested. It may be easier to see the abstracting character of hypnosis in terms of the corresponding time orientation. Although hypnosis may be self-induced, the process for intentionally inducing hypnosis is first learned in the process of being hypnotized by another person and being given the post-hypnotic suggestion that the subject will be able to reproduce the state. Thus the nature of the influence is interpersonally “far.” Dissonance modulation is near because it is perceived as an internal process. It is concrete because it can be resolved into discrete discrepancies. (Dissonance supplies a metric for overall coherence based on discrete discrepancies, a topic to be visited in my Juridical Coherence blog.)

The Role of Suggestion in Persuasive Writing: “What is Classic Prose?” Revisited

Thomas and Turner (Clear and Simple as the Truth: Writing Classic Prose) describe classic prose as nonargumentative, a characteristic both paradoxical and central. Paradoxical because another characteristic of classic prose is advocacy of a thesis, which is usually explicit. (Thomas and Turner conflate introductory exercises involving presentation with full-blooded classic prose, which is fundamentally a tool of argument.) Yet the “nonargumentative” quality of classic prose, the absence of opinionation, pleading and pressure, is unmistakeable. I contend most central in classic prose is its indifference to suggestion: refusing either to wield it as a weapon or compulsively avoid it.

Suggestion is one of the two methods by which writers can influence readers’ far-mode beliefs, the other being cognitive-dissonance modulation. Most of the popular advice about persuasion involves suggestion, my main evidence concerning communication conducive to suggestion coming from the study of hypnosis and of interaction ritual chains, both enhanced by authority, simplicity, brevity, ease, repetition, and emotional involvement. The maxims of plain-talk writing amount to a guide to the forms for persuading by suggestion.

The mentioned prototypes for suggestion also demonstrate the limitation of suggestion as a tool for persuasion. Both hypnosis and interaction ritual chains require prior commitment. Hypnosis doesn’t work against a subject’s will, and confidence in the hypnotist is one of the most important determinants of trance induction. As for interaction rituals, we see today how political rallies excite only the party’s adherents.

The reason for the limitations of communication based on suggestion is the phenomenon of reactance. Attempts at suggestion against a person’s will arouses an opposing resistance often stronger than the suggestive effect, so that the target of the communication moves, by “reverse psychology,” in the opposite direction. Today’s political polarization is associated with reliance on suggestion in campaigns: how many “We’re with her” buttons can nonsupporters see before they start hating her?

Because of reactance, writers attempting to persuade the unreceptive must forswear suggestion. That includes not only avoiding the fallacies of suggestion, such as appeals to authority, assurances of sincerity and credibility (“believe me!”), and vagueness and ambiguity, but also the formal characteristics of suggestion when embodied in prose, the plain-talk techniques developed by advertising specialists: short sentences, common words, repetition (“tell them what you are going to say, say it, and say what you’ve said”)—all of which characterize hypnotic induction. By forswearing suggestion, classic prose attempts to be maximally persuasive while avoiding reactance.

There is another stylistic approach to avoiding reactance actually more extreme than forswearing intentional suggestion: striving to eliminate as much suggestion as possible. The norms and practices of “academese” express this drive to avoid suggestive content, being a systematic display of just those forms that would be avoided in hypnotic induction: long sentences, obscure words, and passive voice. (Contrast with classic prose, where sentence length varies to serve as a tool for emphasis, words are chosen for precision, and active voice enjoys a rebuttable presumption.) “Legalese” emerges as a conflicted style. The need to prevent reactance by avoiding suggestion is expressed in the same way as in academese. But the respectability of using suggestion is greater in law than in science, with the consequence that hypnotic-like incantations obtrude, such as doublets and triplets, and other routinized phrases.

To be sure, no communication can banish all suggestion, but, unlike academese, classic prose doesn’t actually try. In fact, a cynic might contend that classic prose makes suggestion acceptable by hiding it under other stylistic effects. Classic prose disdains emotional forcefulness, for example, but classic writing is in fact forceful by virtue of its employment of emphasis through variation.

If classic prose is indifferent to suggestion, that doesn’t mean it is necessarily rational. Cognitive-dissonance modulation, the other method for persuasion, is entirely compatible with irrationality, but suggestion is incompatible with rationality. It is a form of irrational influence, whereas dissonance modulation may be rational, depending on the particularities of the discrepancies. 

Suggestion is irrational because its mechanism is the induced refusal to make a critical evaluation of the communication. The uncriticized belief is accepted as true because of the unity of comprehension and belief.

Applying the maxim that to avoid reactance use dissonance modulation rather than suggestion has an extra wrinkle for legal writing: attorneys are charged with presenting their clients’ sides. Within these bounds, some tactics arousing excess reactance are ill advised, such as obvious opinionation, exhortation, and over-simplification. And while attorneys should say clearly what they want the court to do, the phrasing “the court must…” should probably be avoided.