Teaching Assistants: David Hoffman on Consumers’ Expectations

Expect some posts about the newish Restatement of Consumer Contracts Law, as there was a symposium on that subject, and it is useful to get a sense of the early assessments of the successes and challenges to the success of that work. We start today with David Hoffman’s contribution, Consumers’ Unreasonable Textual Expectations, which is forthcoming the the Harvard Business Law Review but is available for download on SSRN now.

Professor Hoffman’s article epitomizes something that is true of a lot of really useful scholarship. He problematizes language in the Restatement that at first glance seems solid. Once he has called your attention to the problem, you can never see that language in the same light again. In this case, Professor Hoffman casts his gimlet eye upon Restatement of Consumer Contracts (the RCC), at § 4(d), which provides that “standard contract terms are interpreted in a manner that effectuates the reasonable expectations of the consumer.”

Professor Hoffman begins by noting that our usual methods for contract interpretation, which attempt to reconstruct the intention of the parties, are not of much use when it comes to consumer contracts. The language is boilerplate, often adopted from other contracts, and consumers do not read or understand the language. They intend to buy stuff and know that there are terms attached, but they have no intentions with respect to those terms.

There seem to have been two sources for § 4(d): the Restatement 2d’s “notoriously unsuccessful” and largely unmourned § 211 and some insurance cases. One might at this point rejoice that the Reporters of the RCC were boldly suggesting that contra proferentem should apply to consumer contracts of adhesion. That is, so long as a consumer could establish the ambiguity of a contract, the consumer’s reading should prevail against the drafter. However, Professor Hoffman’s reference to “the usual contra proferentem pablum” suggests that he is not on board. Rather, this tale of Hoffman proceeds through a series of questions.

First, how should courts determine that a contract is ambiguous? As indicated above,  because consumer contracts are not read by consumers either before or after formation, the usual contextual approach to interpretation is not available. There is no context in which to place the text. RCC Reporter Omri Ben-Shahar and Lior Strahileveitz have argued that survey data might be helpful, but that technique is likely too expensive and unwieldy. It also collapses the two-step interpretive process (identifying ambiguity; choosing meaning) into one. Professor Hoffman, with his co-author, Yonathan Arbel (left), in their article, Generative Interpretation (reviewed on the Blog here), have suggested that large language models (LLMs) might aid in interpretation.  The LLM approach remains a two-step process. The LLM can establish the existence of ambiguity; it would still be up to the trier of fact to determine which of the possible meanings should prevail.

Professor Hoffman next asks how we should measure reasonable expectations. Here again, Professor Hoffman starts with the survey-based approach but thinks it can be profitably supplemented with the generative interpretation technique that he and Yonathan Arbel have developed, and which is in a consant state of improvement. 

Next, Professor Hoffman asks whether reasonable expectations can operate as a defense to obligations. The RCC comments suggest that reasonable expectations, which might arise from pre-contractual promises or affirmations or from the general commercial environment, can effect construction even without textual ambiguity.* I wonder if the Reporters have in mind something like the Shellenberger and Mathews cases discussed earlier this week. In the former, a purchaser of a dishwasher expected that her warranty guaranteed that she could get her appliance repaired. She got that impression from some contractual language that included some caveats the significance of which a consumer might not appreciate (“covered repairs,” “where applicable”). In the latter, the case foundered on the consumer’s failure to contest charges in writing within 25 days of receipt of the final invoice. Might a reasonable consumer conclude that phone conversations suffice as notice of contestation? Professor Hoffman reiterates his skepticism regarding the possibility of reconstructing consumers’ reasonable expectations with respect to documents they do not read.

Finally, Professor Hoffman asks whether we should ignore consumers’ unreasonable expectations. We tend to assume that consumers will offer self-serving interpretations of legal texts, but research suggests that consumers often take a textualist approach when their ox is not the one that is gored. Scholars may prefer a more normative account of what “reasonable” means — something like what consumers should be entitled to expect, regardless of their actual expectations. The RCC does not go down that path, and Professor Hoffman congratulates them on their commitment to empiricism. Professor Hoffman concludes with the challenge to scholars to “develop a vision of ‘reasonable expectations’ that harmonizes consumer agency and consumer protection.”

* Program note: later this month, we will have a post on the RCC Reporters’ Guide to the RCC. They think that interpretive matters should be informed by the concept of salience. That is, courts should give weight to the sort of information that registers with consumers, much of which is communicated orally. However, dutiful Reporters that they are, the RCC reflects caselaw, and caselaw does not make much use of the concept of salience. The Reporters advocate for its usefulness in their Reporters’ Notes.