Is there a practitioner who hasn’t faced this question? It may be innocently posed by a parent or teacher or perhaps not so innocently thrown down by a hostile colleague untrained in ABA. Given our everyday dialect, it’s understandable why the difference between positive reinforcement and bribery would come up – to the uninitiated, such arrangements could look like pretty much the same thing.
The challenge of how to answer the question was the subject of a recent thread on the Students of Applied Behavior Analysis Facebook page. Along the way, discussion mostly slipped by a fundamental issue that such topics should raise. In addressing the matter of how we accommodate the dislocations between everyday dialect and our technical lingo, we should proceed only with an understanding of the nature of verbal behavior and how it is that we learn to label anything. (For example, see “The meaning of meaning,” posted in August 2015.) This helps avoid the temptation to instead argue for definitional distinctions that preserve our well-learned convictions.
The vagaries of colloquial connotations
On the question of how reinforcement and bribery are different, before we start we should clarify the definition of positive reinforcement, making sure we are consistent with the underlying science. Easy enough, but only then should we turn to the matter of bribery. Lacking the guidance of an experimental literature about bribery, we focus instead on features suggested by its colloquial connotations. For example, we might note that a bribe benefits the person offering it, perhaps more than the person receiving it, and that a bribe often encourages inappropriate behavior. We might argue for a distinction with reinforcement based on the motivation of the person arranging the contingencies, such as the aiming to build durable behavior change with reinforcement versus induce temporary changes with bribes.
Our arguments might quickly prompt rejoinders from others that offer contrary examples. Bribes may lead to benefits not just for the person offering them, but for the person receiving them, and the target behavior may be desirable for all concerned. The intent underlying a bribe may also be broadly constructive, rather than merely situational. Furthermore, reinforcement contingencies can certainly be found that seem to have the same characteristics as bribes, such as benefiting the person arranging them more than the person experiencing them.
In fact, the longer a discussion goes on about the difference between reinforcement and bribery, the more difficult it may be to identify distinctions that everyone would agree with. This might be the result even if the discussion involved only behavior analysts. We might empanel a task force charged with wrestling this issue to the ground, but even if it eventually proposed a formal distinction – and it might not – we would likely discover that everyone else still finds plenty of room for disagreement.
Sources of control
The way out of this dilemma is to realize that the question – “Isn’t this just bribery?” – is itself a trap. Not only does it presuppose that there is – or at least should be – a straightforward answer, it encourages accepting colloquial dialect as proper context for figuring out that answer. This misdirection is hard to notice because we are inclined to accept the constraints of everyday language without realizing what they are or their impact.
References to positive reinforcement are based on the procedures and findings of a mature experimental science. There is little debate among those familiar with this science about what is going on with positive reinforcement, both procedurally and definitionally. The sources of control over our verbal behavior regarding reinforcement are narrow and precise. Bribery, on the other hand, is a vernacular reference and can only be understood as such. Unlike reinforcement, everyday talk about bribery has only a colloquial – that is, highly varied – foundation, and it may be challenging to identify its distinctive features.
Whether the speaker’s history with these two terms is narrow or broad, it is important to remember that both terms are no more than verbal behavior. Although they more or less describe behavior-environment contingencies and their effects, they are only verbal behavior about these contingencies. The real question is not why we talk about particular contingencies as reinforcement or bribery, but whether there are consistent differences in behavior-environment contingencies that warrant two such distinctive references, whatever the terms applied.
For behavior analysts, the criterion for answering this question should be the same as for all scientific vocabulary – the unrelenting standard of utility. For any term, we should judge whether it helps us ask better research questions or improves our clinical capabilities. If so, there may be reason to accept it into our technical vocabulary. If not, we might best avoid treating it as a technical reference.
We long ago settled this matter for reinforcement, but what about bribery? Is there a distinctive type of reinforcement contingency that warrants a unique identification because of its idiosyncratic effects? Do the vagaries of the vernacular reference hide a distinctive behavior-environment relationship that is sufficiently important to justify a specific label? In the legal arena, for example, there are definitional details that encourage consistency in criminal proceedings, but these particulars don’t necessarily identify special features of what are otherwise still just reinforcement contingencies. Other uses of the term lack the consistency of legal formalities and seem traceable only to our individual verbal communities.
In any event, the point of this discourse is not to propose a definition of bribery that distinguishes it from positive reinforcement. Instead of falling for the temptation to accommodate the term by awarding it this formal distinction, our reaction to the question should be to focus on appreciating the complexities of behavior-environment contingencies while remaining open to evidence that might justify formal subcategories under the umbrella of reinforcement.
But how do I answer the question?
Of course, this cautious and intellectual approach begs the everyday challenge from well-meaning clients, unfriendly colleagues, or your uncle wanting to test you at the Thanksgiving dinner table: “Isn’t this just bribery? “ They want a simple answer, delivered in strictly ordinary terms. We know it doesn’t work like that, but what will usually do the trick is explaining in plain English what’s going on with reinforcement and why it is ubiquitous and important. We can incorporate bribery into this explanation in whatever way seems to work under the circumstances, knowing that our audience doesn’t really want us to start getting technical about the nature of verbal behavior and science-based dialects. The objective of the discussion should be to get them to appreciate the nature of reinforcement, whatever the overlap with everyday terms.