Crowdsourcing Rears Its Head Again: My Thoughts on George Triantis’s Working Paper on Improving Contract Quality

Via this post on Legal Informatics Blog, I learned of an article entitled Modularity and Innovation in Contract Design: A New Path for Transactional Legal Practice, 18 Stan. J.L. Bus. & Fin. 177 (2013). It’s by Associate Dean George Triantis of Stanford Law School.

But apparently that article isn’t yet available—the SSRN link (here) in the Legal Informatics Blog post is to a Stanford Law School working paper by Triantis entitled “Improving Contract Quality: Modularity, Technology, and Innovation in Contract Design.” Before I realized that discrepancy, I had waded through the working paper, with its opaque prose and muddled structure. So here, whether you want them or not, are my thoughts on the working paper. (I suspect that the law review article will closely resemble the working paper.)

To the extent I was able to figure it out, here’s the gist of the working paper:

  • Standardization reduces the cost of transactions and makes them more efficient.
  • Stresses in the legal market are pushing standardization upmarket, increasing the proportion of legal work that can be regarded as a commodity service.
  • Innovation in designing contract terms could provide law firms with new opportunities for premium work.
  • But standardization results in lawyers being reluctant to accept novel terms, so innovation requires more than designing new contract terms—you also have to find a way to disseminate new contract terms so that they’re more readily adopted.
  • Structural impediments preclude trade associations, government, or law firms from being engines of innovation.
  • By analogy with open-source software, contract innovation could be driven by technology-assisted peer production [aka “crowdsourcing”].
  • That’s facilitated by treating contract provisions as “modules” that are largely independent of other provisions in a contract.

Triantis’s article prompted the following thoughts:

First, I don’t see how wiki-driven innovative deal terms could provide law firms with new opportunities for premium work. Unless they somehow result in new or more-lucrative transactions, the innovative terms would quickly become the new orthodoxy.

Second, innovative deal terms of the sort described in the article would seem to be limited to big-time M&A deals and financially driven transactions. That would make them a non-issue for other kinds of contracts.

Third, I don’t understand the emphasis that Triantis places on modularity. For example, he describes how “document assembly software offers menus of discrete contract terms as modules that users can mix and match to customize their needs.” That’s not how sophisticated document assembly works. Instead, answering a given question can result in any number of changes throughout the contract. Those changes can include adding or omitting entire sections, but they can also include adding or subtracting words and phrases.

Fourth, Triantis has an narrow view of “quality.” According to him, improved contract quality is simply a matter of coming up with novel design features. He doesn’t acknowledge the systemic shortcomings of traditional contract language. I suggest that roadblocks to coming up with novel provisions are the least of our problems.

Fifth, I don’t share at all Triantis’s confidence that crowdsourcing can make a worthwhile contribution to developing rigorous contract language. Wikis have been with us for more than ten years, but no initiative as open and decentralized as what Triantis proposes has come close to producing contract language with reliable substance and language; see this 2011 post. Instead, we have the chaos of, for example, LinkedIn groups; see this May 2013 post. (As I described in this post, I’ve benefited from crowdsourcing, but of a more controlled sort.) Triantis doesn’t address this record of futility. Even more damning, the most recent post to the Harvard Law School Contracts Wiki was apparently in 2008, and the sole contributors have been Triantis and Barnard. It takes some chutzpah to offer as an exemplar of open collaboration an initiative that is apparently moribund.

And sixth, in citing open-source software, Triantis is comparing apples to oranges. As I understand it, much of open-source software doesn’t involve, as Triantis puts it, “collaboration across a wide community of otherwise unaffiliated developers.” Instead, it’s usually more controlled than that. For example, Triantis cites Android; well, Google has played a big role in Android. So it’s not clear that open-source software offers a good model for open crowdsourcing of contract language.

So although a recent article described the working paper as “compelling,” I don’t think it does much to advance understanding of what’s required to improve contract language and the contract process.

About the author

Ken Adams is the leading authority on how to say clearly whatever you want to say in a contract. He’s author of A Manual of Style for Contract Drafting, and he offers online and in-person training around the world. He’s also chief content officer of LegalSifter, Inc., a company that combines artificial intelligence and expertise to assist with review of contracts.

3 thoughts on “Crowdsourcing Rears Its Head Again: My Thoughts on George Triantis’s Working Paper on Improving Contract Quality”

  1. Ken writes, apropos of modularity: “That’s not how sophisticated document assembly works. Instead, answering a given question can result in any number of changes throughout the contract. Those changes can include adding or omitting entire sections, but they can also include adding or subtracting words and phrases.”

    Ken, I think that’s one of the main reasons that “sophisticated document assembly” has had such a hard time taking off. It’s been my observation that practitioners like to have close control over the exact wording of the document, phrase by phrase and even word by word.

    If I can indulge in a sweeping generalization: Practitioners can get nervous when they don’t see and control the precise changes that the software is making, in real time as the changes are being made. They fear that the software might make an unseen change somewhere in the document that causes a substantive problem, and that they won’t pick up the problem when reviewing the finished document.

    As to modularity, I think George Triantis has it right. But you’re also right that the lack of contributions to his Contract Wiki project speaks volumes. I think the problem isn’t with modularity per se, but with the fact that as a group, transactional lawyers aren’t as motivated by the desire to create a public good, or for peer recognition, as are Wikipedia editors and open-source software developers.

    Reply
    • Six points:

      First, control isn’t an issue if you use the installed version of ContractExpress: it allows users to see the output document while they’re answering the questionnaire, so they can see how their answers affect the document.

      Second, even if one is working with “black box” document assembly, the gains, in terms of speed, consistency, and quality (if it’s done right) greatly outweigh the fact that you have to read through the output document.

      Third, I think there are more compelling obstacles to law firms adopting document assembly, including (1) the difficulty of achieving economies of scale and (2) the billable hour as a disincentive to efficiency.

      Fourth, the issue of control is irrelevant at companies. Once the legal department sets up the system, the users have no say on wording; they just answer the questionnaire.

      Fifth, as I note in my post, I don’t think that development of open-source software is a helpful model.

      And sixth, I don’t think lawyer stinginess comes into it. Creating optimal contract language requires greater expertise than does writing a Wikipedia entry, so the needs and incentives are different.

      Reply

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.