Through the Looking Glass: A BigLaw Analysis of Indemnification Terminology

Last weekend I indulged myself with this post, in which I observed, without surprise, that I’m hardly ever challenged in the marketplace of ideas. Well, if traditionalists aren’t challenging me, what do they contribute to the marketplace of ideas?

I don’t go looking for such stuff, so for this blog post I’m relying on a tip from a reader. To protect the innocent, I’ll call them “Ben Guest.” Ben sent me a link to a BigLaw blog post entitled “What Does ‘Defend, Indemnify and Hold Harmless’ Mean?”

That post refers a lot to what the U.S. states do. They “require,” they “view,” they “use,” they “hold,” they “consider.” But it’s not always clear whether the authors are referring to statutes or caselaw.

More to the point, that BigLaw post offers no advice on how to get around inconsistent interpretations. In effect, it assumes that contract drafters are at the mercy of whatever meaning statutes or caselaw attribute to indemnify, defend, and hold harmless. It says things like “The inherent meaning of ‘hold harmless’ is subject to interpretation.”

This has an air of learned helplessness about it. (See this 2013 post for more about that.) It ignores that drafters and reviewers have the power, have the responsibility, to say what the deal is, so they avoid the confusion that comes with relying on words and phrases with obscure and unsettled meaning. That’s why I’ve long recommended you dispense entirely with hold harmless; see this 2012 post. And it’s why I recommend you not rely on the word defend to determine indemnification procedures; see this 2014 post. Instead, be specific about what indemnification covers. You can find my indemnification language in this 2013 post.

Another feature of that BigLaw post is that it acts as if no scholarship on the subject exists, as if what you need to know is just a summer associate research project away.

Overall, the message implicit in that BigLaw post is analogous to the conclusion I draw from lack of any meaningful commentary on my more challenging scholarship. The reason why people don’t critique my stuff is that they have nothing to say about it. And that BigLaw blog post is the functional equivalent of having nothing to say about indemnification.

That attitude is consistent with, and perhaps a product of, how we all create contracts: by endless copy-and-pasting, on faith, from precedent contracts of questionable quality and relevance. Passivity in copying, passivity in analysis.

Is it fair to single out one BigLaw blog post? This sort of BigLaw product varies in quality, so I’m not suggesting that that BigLaw post is somehow representative of the whole. But it was produced by a national law firm, so it’s worth considering on its merits. Why is that? Because testing ideas offered by others is part of a functioning marketplace of ideas.

Thanks to Ben Guest for the tip.


(Hey, while you’re here, check out the September start dates for the next three series of my online course Drafting Clearer Contracts: Masterclass.)

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.

2 thoughts on “Through the Looking Glass: A BigLaw Analysis of Indemnification Terminology”

  1. “The phase ‘defend, indemnify, and hold harmless’ is found in many, if not most, contracts with liability allocation provisions, across multiple industries.”

    Shouldn’t the second word be “phrase”?

  2. And the third sentence from the article “For example, an Oklahoma statue regarding the interpretation of an indemnity contract . . . .” There’s apparently a statue in Oklahoma that you can visit for guidance on indemnification. Sloppy


Leave a Comment

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