Revisiting “Provided That”

The last time I had anything to say about provided that on this blog was in this 2008 post. Well, today I was woken from my provided that slumber by this post, entitled “Provisos in Contracts,” on the Paper Software blog.

Paper Software is developer of Turner, a Mac-only software with the tagline “Smarter, simpler contract drafting tools.” The guy behind Turner’s design is Ben Whetsell (@benwhetsell), and it’s Ben who wrote the post on provisos.

Ben’s post prompted me to send out the following fanboy tweet:

I tweeted that because Ben’s post exhibited qualities that I’ve seen in other stuff he’s written: sharp presentation, serious attention to detail, thorough research, and a broad perspective. It’s the sort of stuff that encourages me to up my own game.

Now let’s consider Ben’s conclusions.

He says—and I’m paraphrasing—that if provided that is ambiguous in a particular context, you shouldn’t use it, and that if you want to avoid legalisms, you shouldn’t use it, but that it’s not otherwise a problem, and you should feel free to use it or not, as you see fit.

My thoughts:

Ben includes me in the camp that objects to provisos because they’re archaic. Ben says that “Instead of offering reasons to avoid provisos, these statements play on the desire to be on the cutting edge.” But that doesn’t apply to what I’ve said on the subject. (Cue buzzer sound effect.) MSCD doesn’t suggest that provided that by itself is archaic. Instead, it points to “the archaic trappings of the traditional proviso,” namely the funky underlining, the semi-colon, the extraneous however, and the way that two or more provisos are tacked on to a provision like carriages being pulled by a locomotive (usually with further being added to the formula for the second proviso). If something is gratuitously old-fashioned, that’s reason enough to get rid of it.

Regarding the argument that provisos encourage overlong sentences, Ben says that it’s a straw man, because you can create overlong sentences without provided that. But because provisos are traditionally treated as modules to be bolted on to the back of a sentence, they’re particularly conducive to overlong sentences. Eliminating traditional provisos might reduce the temptation to create overlong sentences.

Ben then tackles another objection to provided that. He quotes various commentators, including the following snippet from MSCD: “In other words, using provided that is an imprecise way to signal the relationship between two conjoined contract provisions.” Ben describes this as a “fallacy,” because generally no ambiguity results—the meaning of provisos is generally evident from the proviso, even though it’s always introduced by provided that.

But even if you assume that provided that doesn’t result in ambiguity, using provided that to introduce different kinds of meaning makes readers work harder. It would simplify matters if drafters were to use different signals to introduce whatever follows.

In a related argument, Ben says, regarding cases cited by commentators to show that provided that has different meanings, that “in most of the cited cases, language that contains provided that is held to be unambiguous.” But I don’t care whether a court ultimately decided that a given proviso was unambiguous. That’s because I don’t want to win fights! The fact that contract parties have demonstrated a propensity to fight over provided that and its variants is enough to make me wary of them.

He goes on to say the following:

If the phrase provided that caused problems with any degree of frequency, it seems that I should be tripping over cases, but finding a single case within the last six months in which provided that caused a problem appears to be an impossibility. It seems that however ambiguous provided that might be in theory, in practice it’s used unambiguously (or at least not so ambiguously that the parties end up suing each other).

But I’m not interested in playing the odds. If a usage is suboptimal and I have clearer alternatives, I’ll go with the alternatives.

So I disagree with Ben on all counts: First, the traditional trappings that often accompany provisos are archaic. Second, provisos are conducive to overlong sentences. Third, use of provided that to introduce different kinds of meaning makes readers work harder. And fourth, people fight over provided that. The fix for those problems is to use something other than provided that and its variants.

By the way, having Ben suggest that, among other things, I’m promoting a fallacy is a trivial price to pay for having someone challenge my writings in a detailed way. I don’t get enough of that. I agree with Ben that one sees a lot of bluster in drafting commentary, and I could do with having more people like Ben around to keep me on my toes.

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.

7 thoughts on “Revisiting “Provided That””

  1. My biggest concern with Ben’s excellent analysis is that he is purely applying “common sense” and not taking into account how parties use and abuse contract language in commercial disputes and how and when the courts decide cases on the meaning of contractual language. Unlike you I do care how the courts interpret contract language.

    So, I don’t agree with the Ben’s premise in some of his comments:

    “The problem with this example in the context of Haggard & Kuney’s overall argument is that the three distinct ways in which provided that is used are all readily apparent. There’s no ambiguity here.”

    – Readily apparent to disinterested bystanders who have a good command of the English language, perhaps. Not all business people are good with words, and whether they are good or not, they may prefer to argue a point to get their own way.

    “If the phrase provided that caused problems with any degree of frequency, it seems that I should be
    tripping over cases, but finding a single case within the last six months in which provided that caused a problem appears to be an impossibility.”

    – Every 6 months? To take a mischievous example, the English courts first commented in 1986 on whether there was a difference between best endeavours and reasonable endeavours, and subsequent cases were spaced several years apart. 6 months is a nanosecond in the development of the law.

    I agree that we shouldn’t argue for changing contractual language for the sake of being cutting edge. But equally, I agree with Ken that provisos are often presented in an old-fashioned, lawyerly way. On balance I would get rid of them, partly for this reason and partly because they can be ambiguous and therefore can be used as a weapon in commercial disputes. Making it harder for self-interested parties to take dubious linguistic points should be part of the objective when improving contract language.

    • And decided cases are, of course, always just the tip of the the disputational iceberg. If there is a case every six months (which there may or may not be), that signifies a large number of disputes that never got to court.

  2. Ken:

    I prefer to eliminate all instances of “provided that” because I don’t have time for the deliberation over whether any particular instance could be ambiguous. I know that the construction can be ambiguous and it is easily replaced by something that is not. Detecting when something might be ambiguous is hard work. It requires you to both hypothesize fact patterns you have not yet encountered, assume that your collaborative negotiating partner turns nasty, and then think about how a court would apply language. Maybe I’d like that if I was getting paid by the hour.


    • After reading this blog on ‘provided that”, the following clause

      “ABC agrees to extend the Warranty Period for any such rectified work to a further Twelve (12) months following the Company’s written approval and acceptance of the rectified work, provided that the extended Warranty period for any rectified work shall not extend beyond 24 months from the date of commencement of the primary Warranty Period stated at cc.c.”

      was changed to

      “ABC agrees to extend ……….acceptance of the rectified work. Notwithstanding the foregoing, the extended Warranty period for any rectified work shall not extend beyond 24 months from the date of commencement of the primary Warranty Period stated at cc.c.”

      Did it work..?


      • How about this: “In the case of rectified work, the Warranty Period will end 12 months after the Company accepts that rectified work or 24 months from the beginning of the Warranty Period for that rectified work, whichever is earlier.”

        If you consult MSCD, you’ll see why “agrees that” is always unnecessary; you don’t need both “acceptance” and “approval”; the verb “accept” is better than the abstract noun “acceptance”; and “shall not” is inappropriate, as it’s language of policy. And I suspect that you don’t need the defined term “Warranty Period.”

        • Thanks Ken for the MSCD pointers. Our copy of MSCD is on its way, ETA 3rd of next month.

          On the Warranty Period, lets simulate to test the clause-

          “Warranty Period” (defined term) of the Plant : 12 months from Completion.

          A pump in the Plant is rectified on the last day of the 6th month of the Warranty Period and is accepted on the same day. The Warranty Period for that pump alone is extended by 12 months ( upto 18 months from beginning of Warranty Period) and the Warranty Period for the rest of the Plant ends as defined. On the last day of 17th month, another rectification is carried out on the same pump. This rectification will trigger the 12 month extended Warranty upto 29 months, but we intend to limit it to 24 months.

          I would prefer to have the “Notwithstanding” clause, so as to serve as a signboard that says “NO EXTENSIONS BEYOND 24 months, MISTER”.

          Your views would help.


Leave a Comment

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