No understandings not specified here

« previous post | next post »

Creative Commons is a "nonprofit organization that increases sharing and improves collaboration" by providing "free, easy-to-use legal tools … [that] give everyone from individual creators to large companies and institutions a simple, standardized way to grant copyright permissions to their creative work". In particular, many Open Access journals now use a Creative Commons license, among them the new eLanguage initiative of the Linguistic Society of America.

But yesterday, at the LSA's annual meeting, someone raised a question about the use of Creative Commons licenses for this purpose. The question has an interesting linguistic aspect — I'd describe it as a matter of pragmatic scope — which I thought I'd explain to you here.

if you check out the authors' instructions at any pretty much any journal, you'll generally find a form where authors assert that the work submitted is their own, and not fraudulent or defamatory, and so on. Thus BioMedCentral, an Open Access publisher which puts out more than 250 journals and is owned by Springer Verlag, has an Author Agreement which states

In submitting a research article ('article') to any of the journals published by BioMed Central Ltd ('BioMed Central') I certify that:

1. I am authorized by my co-authors to enter into these arrangements.

2. I warrant, on behalf of myself and my co-authors, that:

a.  the article is original, has not been formally published in any other peer-reviewed journal, is not under consideration by any other journal and does not infringe any existing copyright or any other third party rights;

b.  I am/we are the sole author(s) of the article and have full authority to enter into this agreement and in granting rights to BioMed Central are not in breach of any other obligation. If the law requires that the article be published in the public domain, I/we will notify BioMed Central at the time of submission upon which clauses 3 through 6 inclusive do not apply;

c.  the article contains nothing that is unlawful, libellous, or which would, if published, constitute a breach of contract or of confidence or of commitment given to secrecy;

d.  I/we have taken due care to ensure the integrity of the article. To my/our – and currently accepted scientific – knowledge all statements contained in it purporting to be facts are true and any formula or instruction contained in the article will not, if followed accurately, cause any injury, illness or damage to the user.

It should be obvious why the publishers, editors, and sponsors of a journal would want this kind of protection.

So far so good. But the next statement on the page is "And I agree to the following license agreement", which is asserted to be "Identical to the 'Creative Commons Attribution License'".

The "Creative Commons Attribution License", in turn, contains this clause:

8.e. This License constitutes the entire agreement between the parties with respect to the Work licensed here. There are no understandings, agreements or representations with respect to the Work not specified here. Licensor shall not be bound by any additional provisions that may appear in any communication from You. This License may not be modified without the mutual written agreement of the Licensor and You.

When readers download the published article, they obviously don't see (much less become a party to) any of the stuff that the author has to sign about originality and truth and so on, they just become the 'You' referred to by the Creative Commons license itself, which contains no reference to the assertions that the authors have previously made about originality, non-secrecy, lack of libel and fraud, etc.  But that Creative Commons license therefore appears to state something false, namely that "There are no understandings, agreements or representations with respect to the Work not specified here".

This sentence appears to be false on its face, since when the paper was submitted, the author(s) signed "understandings, agreements, or representations" about originality and on on, which are absolutely "not specified here" in the license agreement that a reader becomes a party to.  And I gather that a legal advisor to the LSA has expressed the concern that as a result, the authors' claims of originality etc. might be nullified, thus undermining the attempt thereby to provide protection for the publisher.

Now, I am most definitely not a lawyer, but I'm assuming that lawyers employed by the likes of Springer have checked this out and decided that it's OK.  And my working hypothesis is that this is because there's  the previous sentence "This License constitutes the entire agreement between the parties with respect to the Work licensed here" creates an implicit restriction that allows the following sentence to be true.

That's because the cited sentence contains an implicit modifier, something like "There are no understandings, agreements or representations [between the Licensor and You] with respect to the Work not specified here". This is true, since it's the publisher and the reader who are parties to the license, with the author(s) not involved at that point.   Or something like that…

This may be a completely routine aspect of contract law, a subject that I know almost nothing about.  But this reinforces my belief that it would do lawyers good to take a couple of semantics courses (and, of course, it would do semanticists an equal amount of good to take a law course or two); and that it would be nice if there were a way to write English that was completely explicit about matters of scope and reference.


  1. Oskar said,

    January 9, 2010 @ 5:18 pm

    I am most certainly not a lawyer either, but I can tell you that it is not only accepted, but even quite common, for open source software and free content to have multiple licenses. For instance, Mozilla Firefox is under three licenses, the Mozilla Public License (MPL), the GNU General Public License (GPL) and the GNU Lesser General Public License (LGPL).

    As for Creative Commons, if you click "edit" on any wikipedia page and look between the big box and the line where you put your edit summary, it states quite clearly:

    You irrevocably agree to release your contributions under the CC-BY-SA 3.0 License and the GFDL.

    CC-BY-SA is of course the Creative Commons Attribution Sharealike License and the GFDL is the GNU Free Documentation Language. This wasn't always the case, wikipedia used to only license its content using the GFDL, and lawyers from Creative Commons were involved in process of switching, and they didn't seem to have a problem either with using two different licenses.

    As I said, I am not a lawyer and I can't begin to answer this question from a legal standpoint, all I know is that using several different licenses for content is very common and accepted in the free content arena.

  2. Neal Goldfarb said,

    January 9, 2010 @ 5:44 pm

    This is a case where discourse structure and pragmatics are more important than sentential semantics.

    To begin with, the second and third sentences of the license provision should be understood as elaborating on the first sentence, which says that the license “constitutes the entire agreement between the parties with respect to the Work licensed here.” So the next sentence—“There are no understandings, agreements or representations with respect to the Work not specified here”—provides further details about the statement that the license constitutes the entire agreement “between the parties.” And in the legal context, the phrase “the parties” would be understood as referring to the parties to the license agreement: the licensor and the licensee. (This understanding is supported by the third sentence, which deals explicitly with the relationship between the licensor and the licensee: “Licensor shall not be bound by any additional provisions that may appear in any communication from You.”)

    On this understanding, the second sentence does not purport to deny the existence of any other agreements between the author and the publisher.

    More broadly, section 8.e of the Creative Commons license is what’s known among lawyers as an “integration clause.” (More specifically, the first three sentences are what make up the integration clause. The last sentence serves a different purpose and would more appropriately be set of in a separate subsection.) The purpose of an integration clause is to ensure that the agreement containing the clause will be covered by the Parol Evidence Rule, which prevents either party to the contract from offering “extrinsic evidence” such as oral testimony to contradict the written terms. The legal effect of an integration clause derives less from the specific terms of the clause as by the fact that the clause is recognized as an integration clause. And as a speech act, an integration clause is performative or commissive, not assertive. By that I mean that the assertion that the integration clause purports to make may be false — there may in fact be an unwritten side agreement — but by triggering the Parol Evidence Rule, the clause effectively makes the side agreement a nullity by preventing it from being introduced into evidence.

    All of which is to suggest that it’s unlikely that a court would accept the argument that the integration clause in the license has any effect on the representations in the Author Agreement.

    (And in case anybody is wondering: No, there is no such thing as the Langue Evidence Rule.)

  3. J.W. Brewer said,

    January 9, 2010 @ 7:25 pm

    To take it one step further, for the license to be worth anything it tacitly presupposes some sort of prior "understanding" or "agreement" external to the license between the Original Author and the Licensor, without which the Licensor would not be in a position, from a copyright POV, to offer the license. A reading of the integration clause as representing that no such understanding or agreement between the Licensor and the Original Author exists would imply that "You" should conclude that the Licensor probably has no rights to the work and You therefore can't rely on the license as a defense against a potential infringement claim by the Original Author. Since that would be a really dumb and perverse result, it is highly unlikely that a court would so read the license. I'm actually modestly surpised that the license doesn't seem to explicitly contain a form representation from Licensor to You that Licensor has obtained the right to offer the license directly or indirectly from the Original Author, but this is not my area of expertise and maybe such a rep is treated as implicit by the custom of the trade among those who draft copyright licenses or maybe for licenses like these you get what you pay for.

  4. Peter Taylor said,

    January 9, 2010 @ 8:27 pm

    What does

    And I agree to the following license agreement.

    mean? The obvious reading seems to me to be "I agree to license the article to BioMed Central under the terms of the following licence agreement", in which case 8.e is false and there's no getting around it by restricting the scope to agreements between licensor (the original authors, who warrant authorship, legality, good faith, etc) and licensee (BioMed Central).

  5. J.W. Brewer said,

    January 9, 2010 @ 10:30 pm

    To my eye, the only reading of "I agree to the following license agreement" that doesn't fairly rapidly lead to absurd results is that which could be glossed as something like "I authorize BioMed Central to license the article in question on the terms set forth in the following form of license agreement to anyone meeting the defintion of 'You' set forth therein, with BioMed Central playing the role of "Licensor" as set forth therein and taking the appropriate steps to assure that licensees are made aware that I am the "Original Author" as set forth therein." I would therefore consider that the "obvious' reading, but putting that in linguistics rather than lawyer jargon I would agree with Neal Goldfarb that what matters is "discourse structure and pragmatics" meaning you have to read in context.and the context is not sentence-level semantics in a vacuum. Reading in context requires reading each sentence in the context of the document as a whole as well as reading both sentence and document against a background of culture- or subculture-specific knowledge about who is likely to be entering into this sort of agreement with whom for what sorts of purposes, probably including some degree of knowledge of what the negative consequences could be, copyright-law-wise, if various sorts of people did various sorts of unauthorized things with respect to the copyrighted work in question. .

  6. Evelyn said,

    January 10, 2010 @ 2:07 am

    Oh, come on, this is easy: just write all contracts in Lojban. Problem solved.

  7. Stephen Nicholson said,

    January 10, 2010 @ 3:25 am

    This hilariously interesting to me. What Neal said is correct, but let me elaborate.

    Contract is predicated on the existence of parties who negotiate the terms of the contract and assumes people on equal footing are parties. It hasn't fully caught-up with things like shrink wrap licenses.

    What that means is that oral contracts are worth more than the paper they are written on. So, when two people enter into an oral agreement, it can be written down. But the paper with the words on it isn't the contract, the oral agreement is. That can be a problem is there is a dispute over the terms of the contract. One party will often produce the writing and say that those are the therms of the contract, the other party will say that isn't what was agreed to orally and that the oral agreement is the contract.

    To prevent this kind of situation, the person who actually puts the contract on the paper will put in an integration (or merger) clause. The idea is to get the parties to agree that all the terms of the contract on the paper and whatever oral agreements were made are included. (I.e., all the oral agreements have been integrated, or merged, into the the written document.)

    An integration clause triggers, as Neal says, the Parol Evidence Rule. The parol evidence rule says that the court will only look at the paper contract to determine what the terms of the contract are. The court won't use oral testimony to determine what the terms of the contract are.

    As you can probably see, lawyers put integration clauses in all the contracts that they draft. I suspect that the language you're talking about: "there are no understandings, agreements or representations with respect to the Work not specified here" wouldn't ever have been thought to be a warranty that the publisher hadn't entered into a contract with the publisher.

    For example, if I enter into a contract to buy a hundred computers from Apple, the contract's integration clause wouldn't be interpreted to nullify any warranty Apple may from Intel. (Or any other parts manufacturer.)

    Part of the problem is that merger clauses were created back in the days when shrink-wrap licenses simple didn't exist. Normally, the clause would specify who the parties are and who the contract applied to. But today, licenses like the creative common license are written to apply to virtually everybody. An open license like the creative commons license is written not even knowing who the licensor is. So you get things like 8e.

    You are completely right about lawyers needing to take a course in semantics. Lawyers often believe in "magic words" that have to be said in order to get a specific outcome. While this is sometimes true, it often leads to overly wordy contracts with lots of boiler plate that are hard to read. The ironic part is these kinds of arcane wording may lead to more confusion than a more straight forward structor.

  8. Anon said,

    January 10, 2010 @ 5:18 am

    One problem is that legal drafting is not designed to be clear, and this is especially the case in the US, which has not embraced 'plain english drafting', as we like to call it. Legal language is designed to be able to be enforced by a court, and this means that it puts an emphasis on understanding the court's precedents in its language. This often makes things confusing to a non-lawyer (often to lawyers as well), but that's just the way things are (or as a cynic may say, it's so that you need to see another lawyer to understand what the first lawyer wrote!).

  9. Richard said,

    January 10, 2010 @ 7:11 am


    I would never defend ignorance of the law (etc.) as a potential defence in any case. But if someone tries in good faith to find out what regulations apply, to what extent is it just or fair for them to be found to have failed to comply with terms (conditions, laws, etc.) that not only they are unable to understand but that not even the courts themselves can agree on how to construe/interpret?

    Legal drafting should be clear at least to somebody (or some body) i.e. lawyers and courts, and the problem seems too often to be that it fails in this respect and is intelligible to no one.

    It would be highly desirable if such material could, without loss of clarity in this first respect, also be written in a way that is intelligible to the parties to it (whether the public in general in the case of criminal law etc., or the parties to particular contracts in private matters). It seems to me to be (a) an implicit and embarrassing admission of defeat by supposedly intelligent lawyers that they are incapable of doing this and (b) prima facie unfair to expect compliance with strictures whose force is at best hard to discern and at worst ambiguous or self-contradictory. However, this would put quite a lot of lawyers out of business, and perhaps it is impossible to achieve both legal clarity and intelligbility to the parties concerned.

  10. CIngram said,

    January 10, 2010 @ 9:03 am

    Since all this, fascinating though it is, is giving me a headache, can I ask a simpler question on another point of the language of the agreement?

    2. I warrant, on behalf of myself and my co-authors, that:

    a. the article is original, has not been formally published in any other peer-reviewed journal, is not under consideration by any other journal and does not infringe any existing copyright or any other third party rights;

    Is it not possible that that 'formally', should be 'formerly'? and if not, what does 'formally' mean here?

    [(myl) I believe that "formally published" in this case means "not just sent to a few friends" and possibly also "not just put up on my web site".]

  11. Obiter Dictum said,

    January 10, 2010 @ 12:38 pm

    Come on guys.

    1. Publishing agreements are often pretty mediocre, probably because there isn't enough money at stake to justify expensive lawyer time. With knobs on, in the case of academic publishing.

    2. I don't know who is responsible for drafting the Creative Commons licenses but I suspect they may be a committee effort, thus tending to decrease the quality of the drafting.

    3. Clearly the author agreement and the Creative Commons license are used between different parties and for different purposes. Some quick fixes of the inconsistencies have been mentioned above.

    Drawing linguistic conclusions about contract drafting from some idle reading of standard-form publishing agreements seems a pretty poor use of your time. Go out and get some fresh air, or watch Oprah or something.

    [(myl) Thanks for the suggestions! And in return, I suggest that while we're watching Oprah, you might work on your reading comprehension skills. Academic or not, this is a real-world issue, brought up by the legal advisor to a real-world organization with around 5,000 members, and the question at issue is whether the license under discussion does or does not have a flaw serious enough to prevent it from being used. If there is such a flaw, then it also affects literally hundreds of other organization, some of them (like Springer Verlag) much larger ones. And the issue of defamation lawsuits against scholarly or scientific authors and publications is, alas, not at all an academic question.

    "Quick fixes" may indeed be available, but if they are needed, there's good reason to want them to be made in a new version of the CC license(s), and not by having individual users modify individual versions of those licenses.

    My belief from the beginning has been that fixes are not needed, because the wording in question, though perhaps not as clear as it might be, is in fact safe enough from a legal point of view, for the reasons discussed in the body of the post. The other comments tend to support that conclusion. (Of course, the concerned parties will get proper legal advice on the matter.)

    It's true that the main purpose of this post is to promote a somewhat abstract discussion of the interpretive issues involved, from the perspective of linguists as well as lawyers — but these issues seem to me to be interesting ones. Your opinion that publishing contracts are especially poor examples of legal drafting is a a contribution to that discussion — but I wonder whether others will agree that the standards are uniformly higher in other areas.]

  12. Ken Adams said,

    January 10, 2010 @ 1:40 pm

    I agree with Neal as to the source of Mark's uncertainty.

    I never cease to be taken aback at the dysfunctional nature of contract prose. To note just one problem in BioMed Central's author agreement, I like the way it asks the author to certify that the author is warranting as to certain facts. I've written at length on verbs used to introduce assertions of fact in contracts; suffice to say that here I'd use neither "certify" nor "warrant" but instead would use an entirely different structure.

    And the Creative Commons license is no thing of beauty. Besides omitting "between the parties," it exhibits the overuse of "shall" that is one of the hallmarks of mainstream contract language. And the "mutual" in "mutual written agreement" is redundant, as reciprocity is inherent in the notion of agreement. I'd have thought that an ostensibly forward-thinking organization like Creative Commons would have tried a little harder to streamline what it calls its "legal code."

  13. Graeme said,

    January 11, 2010 @ 6:05 am

    The subtext of all this is a cat and mouse game between drafters of legal documents and the courts. Specifically 'activist' common law judges who might look for loopholes to do justice in a particular case; especially where a standard form agreeement is imposed by an organisation on a less powerful and relatively diffuse group such as authors. And that tussle also occurs against a backdrop of various rules, both substantive and procedural, that guide the reading of classes of agreements.

    With that in mind, calls to draft documents in some 'plain English', pellucid to Everyman or at least Everybusinessperson, are naive rather than heroic and imply a law concerned only with certainty rather than a broader concept of justice.

  14. Ginger Yellow said,

    January 11, 2010 @ 9:20 am

    "To my/our – and currently accepted scientific – knowledge all statements contained in it purporting to be facts are true and any formula or instruction contained in the article will not, if followed accurately, cause any injury, illness or damage to the user.

    So how do you publish write-ups of trials of toxic substances?

  15. Bob Lieblich said,

    January 11, 2010 @ 9:56 am

    It's perhaps worth mentioning that the parol evidence rule is itself full of holes. Although neither party may introduce evidence to contradict the terms of the written agreement (assuming the presence of either an integration clause or some other clear indication that the writing is the entire agreement), there are still matters as to which evidence can be taken. For example, a given word in a given agreement may have a meaning on which both parties agreed orally, but that mutual agreement may be unclear from the document. If the parties, in the course of working out the contract language, agreed that "glory" as used in the written agreement would mean "a real knock-down argument," either party could offer testimony to that effect. (Whether the court would believe that evidence is a separate issue. And of course the definition ought to appear in the writing. But even lawyers are human.)

    The most obvious exception to the parol evidence rule occurs when the contract language is ambiguous. A simple example might be "year" as used in a US government contract. The lay person might assume that that means "calendar year" (1/1 through 12/31), but it could also mean "federal fiscal year" (10/1 through 9/30). If a choice between meanings was necessary to a decision, and if the written contract did not otherwise clarify, the court might accept testimony on the issue.

    Finally, there are some contract provisions that simply make no sense as written, even within the context of the entire document. I've encountered a few in my time, and have even sat as a sort-of judge in cases where such language had to be interpreted. In such a case, testimony as to the background and extrinsic context of the provision can be very helpful. Indeed, one of the touchstones of contract interpretation is the context in which the language was chosen and the purpose intended by the drafter to be accomplished. I'd say in the particular case MYL poses, a consideration of the respective interests to be protected will lead to a resolution as to the meaning of what appears in print. And in this case I think the intent is fairly clear from the language of the provision. Ultimately, whether it's clear enough for a particular judge in a particular case can only be resolved by litigation.

    Boy, there's nothing like writing about legal issues to turn language pompous and abstract. Sorry about that.

  16. Jens Fiederer said,

    January 11, 2010 @ 10:42 am

    The sensible thing to do is to turn this whole thread over to the good people at The Volokh Conspiracy ( ).

  17. Thinh Nguyen said,

    January 11, 2010 @ 12:21 pm

    Matt Cockerill asked me to comment on this thread, which is very interesting. I agree with Mr. Goldfarb's analysis. The integration clause in question is considered a "boilerplate" provision and is typically found in many contracts. It is, as he notes, a common rule of construction introduced in contracts in order to cope with some of the more abusive practices relating to the Parol Evidence Rule. Among its uses is to prevent parties to a negotiation from raising statements made during the course of that negotiation to vary the terms of the final agreement after the fact. The intent is that only statements contained in the "four corners" of document are admissible into evidence for the purpose of interpreting the contract.

    In my opinion, this provision should be interpreted in light of this common practice. As Mr. Goldfarb notes, the Publisher Agreement is a distinct and separate agreement that exists between the Publisher and the author, whereas the Creative Commons license is the instrument by which a license is granted by the Author to the public. Therefore, the integration sentence of the Creative Commons license should not be construed to affect the validity of the Publisher Agreement or the other way around.

    The mere fact that the Publisher Agreement references the Creative Commons license and obtains the author's consent to license his or her work under such terms does not make them into one and the same instrument. They remain conceptually separate and distinct agreements, between different parties, and addressing different subject matter, and therefore, each may contain its own integration clause without introducing any necessary logical conflict. In particular, the representations regarding originality contained in the Publisher Agreement may be enforceable by the Publisher against the author, but no such representation is enforceable by others (including the public) under the Creative Commons license (for there are none), nor may such representations be later inferred from extrinsic evidence.

    This opinion should not be taken as legal advice. In important matters, please consult your attorney.

  18. Anon said,

    January 11, 2010 @ 7:43 pm

    Richard, in case you're still reading this thread, your comment reminded me of this article – . In short, some of the best lawyers in the world, and the Delaware courts, couldn't work out the meaning of what they had written. Generally it isn't this bad, but something things are just complicated.

    Anyway, I think that we're speaking at cross purposes here. I only really speak from a corporate law background, but generally lawyers don't want to write things in a way that is 'too clear'. The reason is not so that they are confusing, but rather because they want to cover off a whole range of possibilities. For example, a clause like the following (which I've simplified (!) from a contract I'm looking at right now):

    "The total aggregate liability of [A] to [B] with respect to any and all claims arising out of the performance or non-performance of [building work], whether based in contract, tort (including negligence in whole or in part), strict liability or otherwise, shall not exceed in the aggregate one hundred percent (100%) of the [price of the contract]"

    This may seem to have a lot of unnecessary words and overlap, but it's all necessary. For example, why say "total aggregate liability" instead of "total liability"? Because some clever lawyer is going to get up and say that the parties meant liability per incident if you don't include the word "aggregate". Another example – why "performance or non-performance" instead of performance, which one would presume included non-performance? Because some lawyer once ran an argument that they were separate and that the parties didn't intend to include non-performance in performance if it doesn't explicitly say so, and a judge accepted it.

    Essentially, lawyer write the way they do because their audience isn't the general population – it's the court. Think of it like this – you'd prefer to have a document that used bizarre English but was actually protecting your rights rather than a document that you understood perfectly but had a million loopholes built in.

  19. Anon said,

    January 11, 2010 @ 8:04 pm

    Sorry the link didn't paste properly – it is

  20. AdamsDrafting » Blog Archive » When Linguists Talk About Contract Language said,

    January 12, 2010 @ 10:33 pm

    […] want to see how a different online ecosystem approaches the subject, I suggest you have a look at this post on Language Log and the related […]

  21. Richard said,

    January 13, 2010 @ 3:33 pm


    This is a fascinating debate and one where a middle ground is probably most sensible, so I'm rather adopting the position of devil's advocate for the non-lawyer here (oh, the irony!).

    I understand the lawyer's perspective, and it does seem on the face of it a reasonable one, but practically, in terms of ensuring people behave justly and comply with contracts (rather than the much narrower application to resolution of disputes), this approach may not be so adequate, since it logically leads to unnecessary litigation that could have been avoided by making the terms comprehensible to the relevant parties in the first place. In other words, lawyers may think their audience is only the court but this fails to take account of an important factor, namely preventing things coming before the courts. Avoiding confrontation was probably the original aim of drafting a document in the first place, whether it be a law against some form of violent behaviour or a contract to supply goods or services.

    If the expectation (and intention) behind making laws or contracts at all is that – for the most part – most parties will comply with what they are expected to comply with, then for the law-user litigation is expected to be the exception not the rule (the exact reverse of the lawyer's perspective).

    Hence my observation was intended to be the following: that given the major use of laws and contracts is the 'silent' compliance that does not involve lawyers and perhaps goes unnoticed by them, the audience for at least some legal drafting is and should be those who are required to comply with the terms stated. If they can't understand them, their compliance will only be a matter of chance.

    So, my first preference would be to have a document – whether bizarre or otherwise – that means that my rights won't be violated, i.e. one which brings it about that I won't have to go to law as a result of my rights having been violated (especially by innocent parties who could never have realised they were violating them). Ensuring appropriate redress following a violation comes only a desirable second and certainly ahead of unenforceable loopholes, but it is second to avoiding the situation ever arising. It's the difference between stopping the assassin's bullet and catching him after the victim has been gunned down: I know which I'd prefer.

  22. Thinh Nguyen said,

    January 14, 2010 @ 12:51 pm

    English law, from which Americans inherit our jurisprudence, was very formalistic, and many complaints were thrown out due to technical faults in the form of the pleadings (i.e., the right words were not used). This is also applied to the court's view of contracts: there were many formulas or "magic words" that had to be recited in order to have the desired legal effect. In the latter part of the 20th century, legal reformers (mostly American jurists) tried to simplify legal language, and this resulted in a number of reforms include "plain pleadings," which required courts to do their best to interpret language that was not expressed formally but in laymen's terms. However, this gave courts wider discretion to interpret contracts in order to "do substantial justice." In those situations where a strict interpretation of the language of a contract might have a result that the courts deemed "unjust," a court might find a loophole, or a creative interpretation of the plain language, in order to arrive at the result it deems just. However, this decision, originally intended to do justice in a particular situation, ends up setting a precedent. So, lawyers who want a different outcome for their client, would learn from this case and try to get around the court's reasoning by contorting the language of the contracts they draft. But the next time the courts encounter another case where the language of the contract is contrary to the court's vision of justice, it will find yet another way to invalidate the language or another "loophole", and this leads to ever greater contortions by lawyers drafting on behalf of their clients to address the court's new precedent.

    Ultimately, this tension between trying to achieve legal certainty and trying to achieve justice is at the heart of the problem. If you argue for legal certainty (and formalistic drafting), then you will make it harder to avoid manifest injustice in individual cases. But if you prefer to allow courts to do justice in individual cases, then legal certainty suffers, and parties to a negotiation find it hard to predict the consequences that flow from their choice of language. These are competing values, so it is never obvious which approach is always superior at all times. Rather, the courts tend to swing back and forth between these two views, like a pendulum.

    I would be interested in what linguists think about this or whether there are parallel linguistic problems in other fields.

    One legal commentator has called this the conflict between "crystal rules" and "mud rules." For a very interesting read, see:

  23. Anon said,

    January 14, 2010 @ 7:44 pm

    I'll go against the crowd (and even my hero Ken Adams) and argue that "There are no understandings, agreements or representations with respect to the Work not specified here" is not inconsistent with the existence of the author agreement. This is because we're dealing with two different sets of parties.

    The author agreement is between the author and the publisher. Whereas the license is between the publisher and the end user. The license may indeed constitute the entire agreement *between the publisher and the end user,* even if other agreements concerning the work exist between the publisher and other parties, such as the author.

    But I see a different problem in the author agreement. It says "And I agree to the following license agreement." Presumably, the intent is that the author grants permission to the publisher to distribute the work under the CC license. But that's not clear from the wording of the author agreement. "I agree to the following license agreement" means that I am becoming a party to the license agreement, not that I'm authorizing the publisher and an anonymous end user to enter into the agreement.

    It might have been better if the author agreement had said: "I authorize BioMed Central to distribute the article under the Creative Commons Attribution License, Version (whatever version number)."

  24. Andrew Raybould said,

    January 14, 2010 @ 11:53 pm

    I am almost certain that your final-sentence wish, for ‘a way to write English that was completely explicit about matters of scope and reference’, has been ruled out by Godel. I don’t think we can consider semantics without involving logic: for example, in the case you present here, the issue is whether there is a consistent interpretation of the various statements within the license.

    I think the only place for such a language to hide from Godel is in the corner occupied by those formal systems that are not powerful enough to get into trouble. Propositional calculus is one such system, but if we were to follow its example, then, at best, we would be limited to enumerating judgments on specific cases, without any hope of setting down general rules for deciding issues not yet explicitly settled. This, of course, would be of no use at all.

    In logic, once you admit any contradiction, no matter how trivial, you can prove anything you like, and also its inverse. This is way too fragile for the law, so it takes a different approach, that of having mechanisms and conventions for working around the inevitable contradictions. In fact, these seem to comprise a preponderance of its rules, perhaps trumping even logic, and there seems to be a consensus that both consistency and ethics should be followed (when they are at odds, it gets interesting.) For all the layers of civility cloaking modern civics, however, the ultimate truth remains the same: victory goes to those with the means and will to enforce their preferred outcome (unless chance and the physical world intervene…)

  25. Matthew Cockerill said,

    January 18, 2010 @ 8:29 am

    Anon says:

    But I see a different problem in the author agreement. It says "And I agree to the following license agreement." Presumably, the intent is that the author grants permission to the publisher to distribute the work under the CC license. But that's not clear from the wording of the author agreement. "I agree to the following license agreement" means that I am becoming a party to the license agreement, not that I'm authorizing the publisher and an anonymous end user to enter into the agreement.

    It might have been better if the author agreement had said: "I authorize BioMed Central to distribute the article under the Creative Commons Attribution License, Version (whatever version number)."

    Actually, the agreement signed by the author is one in which they, as copyright holder, license the article for universal distribution. This universal license gives BioMed Central as publisher the rights it needs in order to publish and redistribute the article. The author/copyright holder is licensing the right to redistribution not only to the publisher, but also (in one and the same license agreement) directly to the end-user.

RSS feed for comments on this post