Empiricism and Privacy Policies in the Restatement of Consumer Contract Law and The Faulty Foundation of the Draft Restatement of Consumer Contracts, by Steven O. Weise

by Guest Blogger — Friday, Mar. 22, 2019

“Empiricism and Privacy Policies in the Restatement of Consumer Contract Law” (Empiricism) asks the wrong question and takes the wrong approach to answering that question. A second article in same issue of the Journal, “The Faulty Foundation of the Draft Restatement of Consumer Contracts” (Faulty Foundation) has similar flaws.[1]  Both articles misconceive and overstate the role of “counting” in the preparation of the Restatement, as the Reporters emphasize that the Restatement follows the traditional  ALI approach, which is based on a broad review of court decisions.[2]

In addition, Empiricism unfortunately focuses on and discusses at length the wording of a particular Comment from an earlier draft of the Restatement, which the Reporters clarified and significantly modified in response to comments provided to the Reporters.  Those modifications (described below) were made and published before the editing of the article was completed. Thus, that article extensively criticizes a Comment that no longer exists.

Finally, the real question, as discussed below, is to discern where the law is and where it is heading.  The precise favorable ratio of decisions finding that contracts have been or can be formed through a clickwrap or similar process does not change the result.  The decisions overwhelmingly come to common conclusions on the issues addressed in these articles and the Restatement expresses those conclusions.

The “right” question for consideration when preparing a Restatement is to identify where the law is and where is going

The right question

Each of the articles argues that the set of decisions examined by the Draft includes decisions that are not relevant or have limited relevance and, on that basis, significantly pares the dataset used by the Reporters.  Faulty Foundation concludes that the the smaller dataset considered in that article produces conclusions that are “directionally the same”[3] as those of the Reporters.  Both articles conclude that their examination of the decisions produce results that are less persuasive because of the smaller datasets used in the articles.

The articles exclude or devalue decisions that to not have a holding that a contract exists (as well as for other reasons, which are discussed below).  But this is not the “right” question for purposes of identifying what the law is and whether a trend in the law exists.  The “right” question is not whether, in a particular decision, the court in the end held that a contract existed and enforced the contract. The “right” question is whether, had the party seeking to enforce the contract done everything properly, the court would have found that a contract exists and enforced the contract. The vast majority of decisions indicate that an online contract (using clickwrap or a similar process) can be formed — if the business uses a correct process.

For example, in Faulty Foundation the text associated with footnote 63 states:

In contrast, our review found that 6 cases (11%) did not enforce the clickwrap term.… Accordingly, there is a distinct minority position rejecting enforcement of clickwrap contracts.

The first sentence is correct, but the second is not.  None of the six cases cited “reject[ed]” the use of a clickwrap process to form a contract. In each decision the court indicated that there was a “contract.”  The court declined to enforce a particular term of the contract for other reasons (such as unconscionability).[4]  In short, these decisions are consistent with Restatement, § 2.

Similarly, Faulty Foundation cites Cullinane v. Uber Technologies, Inc.[5] as an opinion that “refused to enforce clickwrap terms.”[6] There the court first observed that online agreements can be formed through a clickwrap process.   The court stated that there is a “two-step inquiry for the enforceability of forum selection clauses in online agreements.” The court held that the particular attempted online agreement before it did not satisfy the first step because a condition for successful formation of an online contract (a reasonable notice requirement) had not been satisfied. The court did not hold that an online agreement could not be formed through a clickwrap process.[7]

Excluded or devalued decisions

The articles provide various additional reasons for excluding or giving less weight to certain decisions:

  • For example, Faulty Foundation suggests that arbitration clauses are “atypical”[8] and that decisions considering arbitration clauses (and forum selection clauses) provide a “poor basis for generalizing rules of contract law.”[9] The courts considering online agreements state expressly that the subject matter of an attempt to form a contract online does not affect the contract formation analysis[10]
  • Faulty Foundation suggests that because Federal court decisions interpreting state law are not binding on state courts, they are a “thin basis” for support;[11] the federal courts considering these matters are careful to state that they are applying state law[12] and state courts often cite the federal decisions

The decisions that the articles consider irrelevant or give little weight to do not undermine the definite trend that the courts find that a contract can be formed through a clickwrap or similar process, if all of the necessary requirements have been satisfied.

The role of counting

Empiricism states that the Reporters “lean heavily on their quantitative study to justify the Draft’s comment on privacy policies.”[13] Faulty Foundation states: “Although the Reporters state that the six empirical studies merely supplement more traditional methods of discerning the ‘DNA of the law,’ those studies are in fact the primary evidence for the draft Restatement’s wholesale revision of the standard rules of contract law in the consumer context.”[14]

As noted in Faulty Foundation,[15] in Council Draft No. 5 the Reporters state that the Restatement is not based on counting.  The Reporters demonstrate that the Restatement is based on a broad review of the case law. The Reporters Notes provide a broad basis of case support for the provisions of the Restatement.

The superseded text

Empiricism focuses on and criticizes a Comment in the Draft Restatement that was significantly clarified well before the final editing of Empiricism. Footnote 1 of Empiricism refers to “Discussion Draft No. 4, 2017”of the Restatement.[16]  The footnote then  quotes a tweet from one of the Reporters dated October 19, 2018.  So the article must still have been subject to editing as of that date.  That footnote goes on to state: “As of the final editing of this Article, the ALI has not published a draft subsequent to the version discussed and quoted herein.”   October 19, 2018 is well after the publication of two additional Drafts, which have the modifications described below.

Empiricism focuses on “whether courts treat business privacy policies as part of their contracts with consumers.”[17] The article quotes[18] from the April 2017 Discussion Draft § 1, Comment 9:

Privacy contracts included. The definition of “Consumer Contract” includes agreements between a consumer and business with respect to the consumer’s personal information, … The rules of contract law, including the specific rules in this Restatement, as well as rules not included in this Restatement, apply to contracts involving personal information.

The quoted language can be read to say that privacy statements are “contracts.”  The intent in that early Draft was to indicate that the rules of contract law, including the provisions of the Restatement, determine whether privacy policies are enforceable as consumer contracts.  But that language does not matter, as it is long gone.

The preparation of a Restatement is an ongoing process.  Drafts are posted and the Reporters have meetings with an ALI Advisers  group and a Members Consultative Group.  ALI members and non-members send comments.  Drafts are discussed at one or more meetings of the ALI Council and the ALI membership (at ALI annual meetings).  The comments (whether from ALI members or others) are posted on the ALI website.  The Reporters take in these comments, review them, and frequently make revisions as suggested by the comments.  Through March 17, 2019 (1:56 PM PDT) there were 131 comments (including comments from nonmembers) posted commenting on drafts of this Restatement — some of them commenting on the Draft’s treatment of privacy policies and counting.

The Reporters, in response to critical comments on treating privacy policies as contracts, clarified and refined the Restatement’s position on this subject and rewrote the relevant Comment in Council Draft No. 4 (December 18, 2017), approximately ten months before an edit to Empiricism.  The edits to the Restatement made clear that privacy policies are “contracts” only “if” they are “contracts” — namely, if the requirements of contract law are satisfied.  The edits removed the “includes” and “apply” language from the April 2017 draft quoted above and added the following to Comment 9:

Specifically, if the data and privacy terms are presented to consumers in a manner that satisfies the definitions of “contract” and “consumer contract,” courts should apply the rules of this Restatement (as well as other contract-law rules, but subject to specific rules of data-privacy law) to determine the formation, scope, and consequences of an agreement over data.

Council Draft 5 (September 18, 2018) (approximately one month before an edit to Empiricism), § 1, Comment 9 further clarified that privacy policies are “contracts” only when they satisfy the requirements for a contract:

… Consumer contracts could govern a variety of topics, including the business’s data and privacy policies. .… Accordingly, this Restatement is not intended to determine which data or privacy terms constitute contracts, but merely to provide principles of contract law for those that do. Specifically, if the data and privacy terms are presented to consumers in a manner that satisfies the definitions of “contract” and “consumer contract,” courts should apply the rules of this Restatement (as well as other contract-law rules, but subject to specific rules of data-privacy law) to determine the formation, scope, and consequences of an agreement over data.

The Restatement does not state that privacy policies are “contracts.”

Empiricism’s focus on the superseded version of § 1, Comment 9 is unfortunate as the clarified version does not say that “consumer contracts” “include[e]” privacy policies, which is a central part of the focus Empiricism criticisms.[19]


In the end, the exact percentage of decisions that conclude that an online contract can be formed by a clickwrap or similar process does not matter — whether it’s roughly 3:1 (in Empiricism) or a higher ratio under the Reporters counting.  Faulty Foundation states that its evaluations of the cases and the evaluation in the Restatement are “directionally the same.”[20]  The trend is overwhelming and there is a complete paucity of decisions that would say that an online agreement (however labeled) cannot be made to work.


Steven O. Weise is a partner in the Corporate Department at Proskauer Rose.

This post is part of a symposium on the Draft Restatement of the Law of Consumer Contracts. All of the posts in this symposium can be viewed here.

[1] Both articles misstate the name of the RestatementEmpiricism refers to it as “Restatement of the Law of Consumer Contracts” (page 46) and Faulty Foundation does the same (page 450).  The actual name, since Preliminary Draft No. 1 (October 28, 2014), is Restatement of the Law, Consumer Contracts.  The insertion of the preposition “of” is not just an idle error — the Restatement does not restate the law “of” consumer contracts because there is no such law.  Rather the Restatement restates the law of contracts as the courts have applied it in the context of consumer contracts, frequently in an online setting.

[2] All emphasis in quoted material is added.

[3] Faulty Foundation, page 464.

[4] The Reporters acknowledge that courts, even though a contract has been formed through an online process (such as clickwrap), sometimes will not enforce the contract because of flaws in the process unrelated to the use of clickwrap.  Restatement § 2, Reporters Notes (“… courts routinely enforce clickwraps absent fraud, unconscionability, or other intervening factors.”).

[5] 893 F.3d 53 (1st Cir. 2018).

[6] Faulty Foundation, note 75.

[7] A few days ago (after the publication date of Faulty Foundation), the First Circuit, referring to Cullinane, stated “The reasonable notice standard has governed online contracts across jurisdictions since the early days of the Internet …”.  See Bekele v Lyft, Inc., 2019 WL 1146759, _ F.3d _ (1st Cir. March 13, 2019).

[8] Faulty Foundation, page 458.

[9] Faulty Foundation, page 463.

[10] See, e.g., Cullinane v. Uber Technologies, Inc., 893 F.3d 53, 61 (1st Cir. 2018) (“While the clauses at issue in Ajemian did not include an arbitration clause, ‘the essential question presented was the same: what level of notice and assent is required in order for a court to enforce an online adhesion contract?’”)

[11] Faulty Foundation, page 463.

[12] See, e.g., Specht v. Netscape Communications Corp., 306 F.3d 17, 27 (2002) (Sotomayor, J.) (“… state law governs the question of whether the parties in the present case entered into an agreement to arbitrate…”)

[13] Empiricism, page 55.

[14] Faulty Foundation, page 451.

[15] Faulty Foundation, pages 465-66.

[16] The only “Discussion Draft” with a 2017 date posted on the ALI website is dated April 17, 2017, for discussion at the American Law Institute 2017 Annual Meeting. This appears to be the draft discussed in Empiricism.

[17] Empiricism, page 47.

[18] Empiricism, page 55.

[19] Faulty Foundation does recognize and discuss the publishing of Council Drafts 4 and 5.  For example, see pagesa 449, 453, 454, 455, 464, 465, and 466.  Empiricism cites Faulty Foundation.  See note 77 of Empiricism.

[20] Faulty Foundation, page 464.

Cite As: Author Name, Title, 36 Yale J. on Reg.: Notice & Comment (date), URL.

This entry was tagged .

One thought on “Empiricism and Privacy Policies in the Restatement of Consumer Contract Law and The Faulty Foundation of the Draft Restatement of Consumer Contracts, by Steven O. Weise

  1. Adam Levitin

    What are we to make of Steven Weise’s “defense” of the draft Restatement? It misframes the legal issue and disregards fundamental rules of American federalism. Tellingly, however, Weise does not defend the Reporters’ error-riddled case coding, and he studiously avoids endorsing their conclusions.

    First, Weise writes:

    “Both articles misconceive and overstate the role of ‘counting’ in the preparation of the Restatement, as the Reporters emphasize that the Restatement follows the traditional ALI approach, which is based on a broad review of court decisions.”
    Yet the first appearance of the Reporters’ claim to have followed a traditional approach is in Council Draft No. 5, after they were aware of the damning criticisms of their empirical approach. Remarkably, they claim that the traditional approach arrives at precisely the same position as case (mis)counting.

    Is this claim credible? Given the massive miscoding, very little of which anyone contests (what of the inclusion of statutory and business contract cases, for example?), why should anyone have confidence that the Reporters have now accurately distilled the law using the traditional approach, which also starts with reading the cases? Weiss is a careful lawyer, and the only evidence he cites in support of the argument that this draft Restatement is the result of a traditional analysis is that “the Reporters emphasize” that it is and that the Reporters “demonstrate that the Restatement is based on a broad review of the case law.” But a Restatement is traditionally based on more than a “broad review of the caselaw”; a traditional analysis requires a careful, thoughtful, and forthright reading of all of the major cases.

    Second, Weise argues that neither our replication study nor Professor Klass’s asks the right question. Weise claims that “The ‘right’ question is whether, had the party seeking to enforce the contract done everything properly, the court would have found that a contract exists and enforced the contract.”

    No, that is not the correct question. There are many good reasons that dicta—what a court might have held under different circumstances—is not law. The right question is “what needs to be done for the court to find an enforceable contract.”

    Weise uses clickwrap as an example and that illustrates the problem with his framing. The issue is not whether clickwrap is _ever_ enforceable. No one suggests that a contract can never be formed through clickwrap, and Weise misreads our article in this regard. Instead, the question is “under what facts and circumstances will a clickwrap contract be enforced?”. The position advanced by the draft Restatement is that clickwrap contracts are enforceable because the consumer has manifested assent by clicking after receiving reasonable notice of the standard contract terms on the website. We show that there is a distinct minority of cases that are not in accord with that position. In fact, the Cullinane case Weise cites to does not merely require “reasonable” notice; it goes further to require that the terms of the contract be “reasonably communicated” and found that because the terms were no more conspicuous than other text on the website, they were insufficiently conspicuous to meet this test. Cullinane v. Uber Techs., Inc., 893 F.3d 53, 64 (1st Cir. 2018).

    And here is the real rub. If the Reporters engaged in the traditional Restatement analysis, they would have to grapple with that minority of cases—which includes Judge Jack Weinstein’s opinion in Gogo, the most detailed and thoughtful opinion in the clickwrap cannon—and explain why the minority position was the inferior position. The Reporters don’t do that. Instead, they all but deny the minority position’s existence based on their case counting methodology. The lack of analysis of the minority position in clickwrap cases gives lie to the claim that the Restatement is based on a traditional, thoughtful and forthright analysis.

    Third, Weise advances a sort of harmless error argument, namely that the replication studies do not challenge the ultimate directional findings of the Reporters. Accuracy is an important virtue in and of itself for a Restatement. Why should someone respect the positions of a Restatement that does not accurately identify the relevant law? The coding errors here were not de minimis. We found that roughly two-thirds of the modification cases and half the clickwrap cases were misunderstood by whomever performed the coding. Even if we were ourselves wrong half the time (and, to be clear, no one contends that we were), the scale of the miscoding would still be enormous.

    To be fair, Weise’s argument is that at most we have winnowed down the set of supporting cases, not challenged the direction of those remaining cases. While we believe that the remaining cases often involve particular hard-to-generalize fact patterns, even if they are directionally correct, there’s just not much case law remaining, and little of it is controlling because it is from either federal courts sitting in diversity jurisdiction or state trial courts. As we observed in our article, there are only 9 published state court decisions dealing with the enforceability of clickwrap and those decisions are from 8 states. Most state courts, including California, which represents 12% of the population, have not taken any position on clickwrap enforceability.

    Weise argues that the lack of controlling law doesn’t matter because federal courts are careful to say that they are applying state law and because state courts “often” cite federal court decisions. This may be true, yet surely an attorney like Weise would never represent to a court that such rulings of federal courts sitting in diversity jurisdiction are controlling law. Nor would he cite to a court Bekele v. Lyft, a case he cites here ,but in which there is no holding regarding assent because “Bekele waived the contract formation issue by not raising it in his opening brief.” 2019 WL 1146759, at *4 (1st Cir. Mar. 13, 2019). So why is he claiming that what isn’t good enough for a lawyer appearing before court is good enough for a Restatement, which lawyers will then cite to courts? A Restatement cannot dispense with the realities of our federalism, but Weise’s approach (much like Clay Gillette’s) would reduce “law” to a fuzzy set of principles from which Reporters are free to pick and choose. If that’s the case, what is the purpose of a Restatement?

    The real lesson from our study is that there is simply not a body of law that can be characterized as “the common law” of consumer contracts other than the traditional principles of assent and protection of the nondrafting party that are already contained in the Restatement, Second, of Contracts. As reflected in many of the cases the Reporters (mis)counted, most consumer law is statutory, making the area an ill fit for a Restatement. Even this project’s defenders do not seriously or substantively defend the current draft.

    –Adam Levitin (on behalf of the authors of The Faulty Foundation of the Draft Restatement of Consumer Contract Law)


Leave a Reply

Your email address will not be published. Required fields are marked *