Grice and beer in Federal Court
« previous post | next post »
The philosopher Herbert Paul Grice was cited in an opinion issued on 5/24/2019 by Judge William M. Conley in the United States District Court for the Western District of Wisconsin.
The case is MillerCoors, LLC v. Anheuser-Busch Companies, LLC (3:19-cv-00218-wmc, and as the judge's opinion explains,
During Super Bowl LIII, defendant Anheuser-Busch Companies, LLC, launched an advertising campaign highlighting plaintiff MillerCoors, LLC’s use of corn syrup in brewing Miller Lite and Coors Light, as compared to Anheuser-Busch’s use of rice in its flagship light beer, Bud Light. This lawsuit followed, with MillerCoors asserting a claim of false advertising under the Lanham Act, 15 U.S.C. § 1125(a)(1)(B).
Grice seems to have entered the case through an argument made by the plaintiff in an injunctive hearing held on 5/16/2019 , for which I haven't been able to get the transcript. Quoting the judge's opinion again,
In finding a literally true statement misleading, courts consistently seem to rely on disparaging or derogatory references to the ingredients in the competitor’s product, e.g., Eli Lilly, 893 F.3d at 379 (depicting rbST as a “cartoon monster with razor sharp horns and electric fur”); […]
When counsel for plaintiff was pressed at the hearing to identify a comparable characteristic in these advertisements, he appeared to rely on the scale of the campaign alone — that is, Anheuser-Busch has produced a lot of advertisements, displayed them during high profile events, and mentioned “corn syrup” repeatedly. Specifically, plaintiff relied on the “maxim of relevance,” explaining that under this theory:
if a speaker says something that is capable of more than one meaning and knows that . . . one of those meanings is something that’s relevant and meaningful information to the listener and that the other meaning is something that is irrelevant and not meaningful to the listener, the listener will assume that the speaker means the meaningful, relevant statement because why would you be telling me something that’s not important.
(Hr’g Tr. (dkt. #56) 12.)16 Intuitively, the argument is appealing: why are these ads repeating “made with” or “brewed with” corn syrup if it does not mean that the corn syrup works its way into the final product?17 Still, plaintiff fails to direct the court to any case, and this court could not find any, relying on the pervasiveness or scale of an advertising campaign to find a neutral, truthful statement misleading within the meaning of the Lanham Act.
16 The maxim of relevance (originally called the “maxim of relation”) is one of philosopher Paul Grice’s four conversational maxims, which jointly constitute the cooperative principle. See H.P. Grice, Logic and Conversation, in 3 Syntax and Semantics: Speech Acts, 47 (P. Cole and J.L. Morgan, eds. 1975).
This all started with commercials broadcast during Super Bowl LIII, the Oscars, etc., like this one:
https://youtu.be/LkHvj_KEHBk
Apparently Grice didn't carry the day, though in the end the judge decided to
grant plaintiff a preliminary injunction, though more narrow in scope than that sought by plaintiff, enjoining defendant’s use of the following statements: (1) Bud Light contains “100% less corn syrup”; (2) Bud Light in direct reference to “no corn syrup” without any reference to “brewed with,” “made with” or “uses”; (3) Miller Lite and/or Coors Light and “corn syrup” without including any reference to “brewed with,” “made with” or “uses”; and (4) describing “corn syrup” as an ingredient “in” the finished product.
For more fun with commercial implicatures, see the opinion, which describes other videos and reproduces some print ads.
[h/t Eugene Volokh]
Philip Taylor said,
May 28, 2019 @ 6:49 am
My initial reaction was "OFFS", as it frequently is in reported cases of commercial litigation, but I am puzzled by one part of the ruling. What exactly did the court mean when it
My interpretation of "enjoining" (supported by the OED) is that it is a positive concept, prescribing a course of action, which in the context of the above would suggest that the court has required [the] defendant to make statements such as "Bud[weiser] Light contains 100% less corn syrup" (of itself strange, since there is no indication as to with what it is being compared) and analogous statements concerning points 2, 3 & 4. But if the court is granting [the] plaintiff an injunction, I would expect it to be ruling that [the] defendant should NOT take some course of action which it had previously been practicing. So what exactly is going on here, and what is meant by this part of the court's ruling ?
Ben Orsatti said,
May 28, 2019 @ 7:13 am
injunction (in-jəngk-shən) n. (16c) A court order commanding or preventing an action. • To get an injunction, the complainant must show that there is no plain, adequate, and complete remedy at law and that an irreparable injury will result unless the relief is granted. — Also termed writ of injunction; equitable injunction. See IRREPARABLE-INJURY RULE.
“In a general sense, every order of a court which commands or forbids is an injunction; but in its accepted legal sense, an injunction is a judicial process or mandate operating in personam by which, upon certain established principles of equity, a party is required to do or refrain from doing a particular thing. An injunction has also been defined as a writ framed according to the circumstances of the case, commanding an act which the court regards as essential to justice, or restraining an act which it esteems contrary to equity and good conscience; as a remedial writ which courts issue for the purpose of enforcing their equity jurisdiction; and as a writ issuing by the order and under the seal of a court of equity.” 1 Howard C. Joyce, A Treatise on the Law Relating to Injunctions § 1, at 2–3 (1909).
– affirmative injunction See mandatory injunction.
– anti-antisuit injunction (1988) An injunction prohibiting a litigant subject to the jurisdiction of a local court from seeking in a foreign court to restrain the continuation of a proceeding in the local court. Cf. antisuit injunction.
– antisuit injunction (1961) An injunction prohibiting a litigant from instituting other, related litigation, usu. between the same parties on the same issues. Cf. anti-antisuit injunction.
– common injunction (18c) Hist. 1. An injunction grantable as an order of course, without reference to the merits, when the defendant failed to appear or failed to timely plead, answer, or demur. 2. English law. An injunction issued by a court of equity forbidding enforcement of a common-law judgment. • In some cases, common law and equity rules differed, which could lead to inconsistent remedies. A court of equity that ensured the equitable rule would prevail by issuing a common injunction. Common injunctions were abolished by the Judicature Act of 1873, § 24(5). — Also termed equitable common injunction.
– equitable common injunction See common injunction.
– ex parte injunction (1854) A preliminary injunction issued after the court has heard from only the moving party. — Also termed temporary restraining order.
– final injunction See permanent injunction.
– headstart injunction (1984) Trade secrets. An injunction prohibiting the defendant from using a trade secret for a period of time equal to the time between the date of the secret's theft and the date when the secret became public. • So named since that period is the “head start” the defendant unfairly gained over the rest of the industry.
– hyperinjunction (2009) English law. Slang. A superinjunction that expressly applies its bar on discussing the subject matter or even the injunction's existence to journalists, Members of Parliament, and lawyers except for the recipient's council. • The first known hyperinjunction was issued in 2006. See GAG ORDER (1). Cf. superinjunction.
– injunction pendente lite See preliminary injunction.
– interlocutory injunction See preliminary injunction.
– mandatory injunction (1843) An injunction that orders an affirmative act or mandates a specified course of conduct. — Also termed affirmative injunction. Cf. prohibitory injunction.
– permanent injunction (1846) An injunction granted after a final hearing on the merits. • Despite its name, a permanent injunction does not necessarily last forever. — Also termed perpetual injunction; final injunction.
– perpetual injunction See permanent injunction.
– preliminary injunction (1828) A temporary injunction issued before or during trial to prevent an irreparable injury from occurring before the court has a chance to decide the case. • A preliminary injunction will be issued only after the defendant receives notice and an opportunity to be heard. — Also termed interlocutory injunction; temporary injunction; provisional injunction; injunction pendente lite. Cf. ex parte injunction; TEMPORARY RESTRAINING ORDER.
– preventive injunction (1882) An injunction designed to prevent a loss or injury in the future. Cf. reparative injunction.
– production injunction (1994) Trade secrets. A permanent injunction prohibiting specified conduct in a field or activity that the court has found to embrace misappropriated trade secrets.
– prohibitory injunction (1843) An injunction that forbids or restrains an act. • This is the most common type of injunction. Cf. mandatory injunction.
– provisional injunction See preliminary injunction.
– quia-timet injunction (kwI-ə tI-mət or kwee-ə tim-et) [Latin “because he fears”] (1913) An injunction granted to prevent an action that has been threatened but has not yet violated the plaintiff's rights. See QUIA TIMET.
– reparative injunction (ri-par-ə-tiv) (1955) An injunction requiring the defendant to restore the plaintiff to the position that the plaintiff occupied before the defendant committed a wrong. Cf. preventive injunction.
– special injunction (18c) Hist. An injunction in which the prohibition of an act is the only relief ultimately sought, as in prevention of waste or nuisance.
– superinjunction (2009) Slang. English law. (2009) A type of gag order that forbids the recipient not just to discuss the subject matter but also to reveal the existence of the injunction itself. • Since Members of Parliament are not restrained by the injunction because of parliamentary privilege, and because parliamentary proceedings may be reported without restriction, the existence and contents of a superinjunction may be made known indirectly. — Also written super-injunction. See GAG ORDER (1). Cf. hyperinjunction.
– temporary injunction See preliminary injunction.
– use injunction (1995) Trade secrets. A permanent injunction prohibiting the use of specified information that the court has found to constitute a trade secret.
INJUNCTION, Black's Law Dictionary (10th ed. 2014)
Jerry Friedman said,
May 28, 2019 @ 8:05 am
Philip Taylor: Here's the American Heritage Dictionary on "enjoin":
1.
a. To direct (a person) to do something; order or urge: The doctor enjoined the patient to walk daily.
b. To require or impose (an action or behavior, for example) with authority and emphasis; prescribe.
2. To prohibit or forbid: The judge enjoined the merger of the firms. The court enjoined the company from merging with its competitor.
I imagine the contradictory sense 2 arose from some sense of "injunction".
Caroline Stinson said,
May 28, 2019 @ 8:05 am
Philip Taylor: what’s going on is that “enjoin“ is a contranym, like “sanction” or “ravel.” It can mean either “encourage” or “forbid.” The American Heritage Dictionary has only the “forbid” sense. I believe the OED documents both senses. In a legal context, especially in the context of a ruling on an injunction, the meaning is unambiguously “forbid.”
Since this is readily discoverable from a quick Google, I suspect that you’re aware of this and that your question was meant to be playful. If so, very cool to point out its contranymitude. I had never recognized it as such before and I love contranyms.
J.W. Brewer said,
May 28, 2019 @ 8:52 am
Hmm. I don't think the noun "injunction" in its legal-jargon sense is a contranym even though a given injunction may (as noted in one of the block quotes above) either require specified action(s) (a "mandatory injunction") or forbid specified action(s) (a "prohibitory injunction"). I realize that it doesn't follow from that that the related verb "enjoin" cannot be a contranym, but I remain dubious that it is. One further ambiguity is that although enjoin is or at least can be transitive, the semantic role of the object can vary. So here "enjoining defendant's use of …" is to a lawyer synonymous with "enjoining defendant from using …" (or "… from the use of …"). I think when the defendant him/her/itself is the syntactic object, the verb has less of a contronym feel, because the core sense is that the defendant is being ordered to comply with the court's directions on pain of being held in contempt if he/she/it does not comply, with whether compliance requires affirmative action or simply refraining from action being a secondary issue.
J.W. Brewer said,
May 28, 2019 @ 9:02 am
On the substance of the issue, while the Gricean point makes sense, I would have thought the more obvious legal tack to take would be to argue by analogy to the doctrine one frequently sees in fraud cases where affirmative statement A, while literally true when read in isolation, may be considered fraudulent if important contextual fact B (such as, here "the substance used in one step of the manufacturing process is not actually present in the finished product") is not disclosed. There is probably a Gricean-style analysis that would explicate the bottom-line conclusion that the hearer/reader seeing A w/o B is likely to assume not-B from the silence and thus be misled, but one typically sees that sort of conclusion just asserted on the basis of common-sense understanding, without the analysis implicit in that understanding being spelled out.
Ben Orsatti said,
May 28, 2019 @ 9:30 am
Well, now, this is interesting. I just went and searched the Westlaw corpus (advanced: (enjoin enjoin! injunct!) & DA(bef 01-01-1800)), dead-certain that I'd discover that enjoin/injunction had _always_ had the negative meaning of "forbid", (i.e. "enjoin from") and never the positive meaning, at least in the law, but lookie here; in the late 18th century, at least, the term seemed to equally encompass both "modes" of meaning.
For example, the second quoted paragraph below may be "simplified" to read something like:
"The law in question was intended to declare that the English King's laws were no longer valid. But, that's not exactly true, because even though the first part of the law nullifies English law, the second part expressly enjoins (i.e. "compels") obedience to English common law."
"Afterwards, we find, that a General Convention, elected by the people, met on the 15th day of July, 1776, for the express purpose of framing a new government; and during the sessions of this body, its members, collectively, assumed the powers of making ordinances, of appointing members of Congress, and of defining High Treason, and its punishment All their proceedings and injunctions, except the ordinance respecting treason, were approved and executed; and the Constitution which they eventually agreed upon, was incontrovertibly a dissolution of the government, as far as related to the powers of Great Britain, but not in relation to the powers which had been before exercised by councils and committees.
[…]
"The act for the revival of the laws, passed the 28th of January, 1777, was intended, I think, merely to declare, that those laws, which were originally enacted under the authority of George the Third, ceased any longer to derive their virtue and validity from that source. But there is great inaccuracy in penning the act; for, though it would seem, by the former part of the second section, to be the sense of the legislature, that from the 11th of May 1776, to the 10th of February, 1777, the operation of all the acts of assembly should be suspended; yet, in the close of the same section, obedience to those acts, to the common law, and to so much of the statute law of England, as have heretofore been in force in Pennsylvania, is, with some exceptions in point of style and form, expressly enjoined. We may, however, fairly infer from the general tenor of the act, that those who framed it, thought the separation from Great Britain worked a dissolution of all government, and that the force, not only of the acts of Assembly, but of the common law and statute law of England, was actually extinguished by that event."
Respublica v. Chapman, 1 U.S. 53, 58, 1 L. Ed. 33 (Pa. 1781)
There are other fun little linguistic nuggets in this opinion, such as the peculiar spelling of "chuses" (<chooses), to a conclusion that includes sentence diagramming!:
"Here, then, the matter rests. Had, the issue been in the disjunctive, the prisoner would clearly have come within the description of an inhabitant of Pennsylvania; but when the word subject is annexed, it means a subjection to some sovereign power, and is not barely connected with the idea of territory it refers to one who owes obedience to the laws, and is entitled to partake of the elections into public office."
Respublica v. Chapman, 1 U.S. 53, 60, 1 L. Ed. 33 (1781)
Ben Orsatti said,
May 28, 2019 @ 9:33 am
Edit to prior post. Strictly speaking, my "paraphrase" should read more like:
"The statute in question was intended to declare that the English King's laws were no longer valid. But, that's not exactly true, because even though the first part of the law nullifies English law, the second part expressly enjoins (i.e. "compels") obedience to English common law and to English statutory law, to the extent Pennsylvania courts have applied it thus far."
Haamu said,
May 28, 2019 @ 10:13 am
Just a hypothesis here, after consulting Garner's Dictionary of Modern Legal Usage on this topic.
There would be nothing contranymic about enjoin if we were to require inclusion of the appropriate prepositional phrase (using either to/upon or from). I suspect that, over the years, legal writers found themselves using the from option so predominantly that it came to be seen as the default sense. I wonder if this is how other apparent contranyms have evolved.
Philip Taylor said,
May 28, 2019 @ 10:23 am
Caroline S — No, nothing playful. I was genuinely unable to parse the court's ruling in any sense that made sense to me. I shall re-consult the OED to see if it also offers the "forbid"/"proscribe" meaning. Ah yes, it does — I did not read far enough :
Ben O — Not entirely clear why the long discourse on "injunction" when it was "enjoining" that was unclear to me. But perhaps your message was unrelated to mine.
J.W. Brewer said,
May 28, 2019 @ 10:57 am
On further reflection, here's what's potentially wrong with the Gricean approach in a false-advertising lawsuit. Commercial advertising routinely flouts one or more Gricean maxims. For example the "water is wet" genre of claim flouts the maxim of relevance, and one could identify similar Gricean problems with many of the other standard techniques listed here: http://home.olemiss.edu/~egjbp/comp/ad-claims.html. Yet we assume (at least for regulatory/legal purposes) that consumers are habituated to many of the misleading styles of advertising rhetoric, and thus for an advertisement to be *so* false and misleading that it might actually get the advertiser in legal trouble something more serious must have happened. Here, of course, what is in issue is an advertiser's negative claim about a competitor's product rather than its positive claim about its own product, which I suppose might make some difference in terms of how much deviation from strict adherence to the Gricean maxims a consumer ought to expect.
ohwilleke said,
May 28, 2019 @ 1:17 pm
"Well, now, this is interesting. I just went and searched the Westlaw corpus (advanced: (enjoin enjoin! injunct!) & DA(bef 01-01-1800)), dead-certain that I'd discover that enjoin/injunction had _always_ had the negative meaning of "forbid", (i.e. "enjoin from") and never the positive meaning, at least in the law, but lookie here; in the late 18th century, at least, the term seemed to equally encompass both "modes" of meaning."
The current meaning of "injunction" in the legal sense in the U.S. is primarily derived from its use in litigation in the English Chancery Court from the mid-17th to the early-19th centuries and from legal treatises reporting and discussing those cases in that time frame.
Ben said,
May 28, 2019 @ 1:21 pm
“Ben O — Not entirely clear why the long discourse on "injunction" when it was "enjoining" that was unclear to me.”
Funny, I’d never wondered about why the gerund and nominalization of the same word were slightly different until you mentioned it. So, I looked it up on etymonline, only to discover that, as with much legal language, one word came directly from Late Latin, and the other, through Norman French, but both ultimately deriving from Latin, “iniungere”, from PIE *yeug, meaning “to join”. I wonder if “zeugma” belongs in that family too?
injunction (n.)
early 15c., from Late Latin iniunctionem (nominative iniunctio) "a command," noun of action from past participle stem of Latin iniungere "impose, inflict, bring upon," literally "attach to," from in- "on" (from PIE root *en "in") + iungere "to join together," from nasalized form of PIE root *yeug- "to join."
enjoin (v.)
c. 1200, engoinen, "to prescribe, impose" (penance, etc.), from stem of Old French enjoindre (12c.) "impose (on), inflict; subject to; assign (to)," from Latin iniungere "to join, fasten, attach;" figuratively "to inflict, to attack, impose," from in- "on" (from PIE root *en "in") + iungere "to join together" (from nasalized form of PIE root *yeug- "to join"). Related: Enjoined; enjoining.
J.W. Brewer said,
May 28, 2019 @ 1:33 pm
Much of English legal jargon comes from French, specifically the "Law French" jargon that came over as a result of the Norman Conquest and persisted for many centuries after the upper classes had stopped speaking Norman French as an L1, but that Law French itself had lots of Latin jargon in it that had never been "domesticated" from Latin into French. I'm not sure if there's a strong pattern as to which bits of jargon were Latin and which French, but its curious (although perhaps not unprecedented) that in this instance we got the verb enjoin as transmitted via French but the related noun injunction in much closer to the Latin version rather than transmuted into something Frenchier. That said, you can find "enjoinder" (the Frenchified parallel you might expect) listed in reference books as a synonym for "injunction" and the google books corpus reveals a handful of instances in which it has actually been used during the 20th century. But I've never personally come across it in the wild (and after over a quarter-century of making a living practicing law, I've probably seen a lot more legal-jargon-heavy texts in the wild than most people).
Michael Watts said,
May 28, 2019 @ 2:11 pm
On the merits, if the facts are these:
And the advertising campaign makes these two claims:
I'd be fine with concluding that this was a literally true but nevertheless legally misleading statement. Most people will not only fail to think of the possibility that Coors Light contains no corn syrup and therefore Bud Light's no corn syrup is technically "100% less" than that; they will actively resist interpreting the claim that way even if you explicitly tell them that that's what the sentence meant.
And on that note…
The special nature of 100% decreases actually means that this sentence can be true regardless of what Bud Light is being compared to. Funny thing.
J.W. Brewer said,
May 28, 2019 @ 10:30 pm
I am belatedly reminded of the young E. Pound's playful use of a catalog of terms that are either archaic-sounding legal jargon or red herrings that sound as if they ought to be.
"But the judges,
Being free of mediaeval scholarship,
Will pay no attention to this,
And there will be only the more confusion,
Replevin, estoppel, espavin and what not."
Philip Taylor said,
May 29, 2019 @ 2:00 am
Michael — ("The special nature of 100% decreases actually means that this sentence can be true regardless of what Bud Light is being compared to. Funny thing"). That is a truly fascinating observation. Despite the fact that I am well over 70 and have always regarded myself as being numerate as well as literate, the true significance of "100% less" completely passed over my head. Thank you so much for pointing it out — I shall be intrigued to find out to how many (if any !) of my seemingly numerate friends your observation is patently obvious.
Adrian said,
May 29, 2019 @ 9:16 am
The US is a topsy-turvy place where a company can even be proud of making beer from rice.
Gregory Kusnick said,
May 29, 2019 @ 10:44 am
I'm skeptical that the writers intended it or that the audience understood it to mean "100% less than zero is still zero". The intended meaning seems clear: "However much corn syrup those other guys use (at any point in their process), that's how much we don't use." Whether or not any corn syrup remains in that form in the final product seems beside the point.
That said, I'm with Adrian that substituting rice for corn in this process scarcely seems like an improvement. And it seems equally odd that any self-respecting brewer would refer to their product as being "manufactured".
Michael Trittipo said,
May 29, 2019 @ 1:41 pm
Thanks for the link to the decision. It led me to wonder about the earliest invocation of Grice by name. The earliest citation by a U.S. court in a published decision seems to have been in 2004: "Implicit assumptions often underlay conversational exchanges, so that a participant in the exchange can communicate more information than what his words would mean in isolation." Lambert v. Blackwell, 387 F.3d 210 (3rd Cir. 2004). It's a bit surprising that there was nothing earlier, given how long the courts have used "canons of construction."
It's certainly true that most uses of "enjoin" in legal decisions involve the sense of prohibition. But by no means all uses. The use of "enjoin" in the sense of "mandate" is long-standing — and still common enough in modern usage that it can be be used without gloss.
Just this year, the intermediate appellate court in my state wrote: "A writ of mandamus 'may be issued to any inferior tribunal, corporation, board, or person to compel the performance of an act which the law specially _enjoins_ as a duty resulting from an office, trust, or station.'" Bergeron v. Comm'r of Corr. (Minn. App. 2019) [_emphasis_ added]. The court saw no need to explain that "enjoins" there is like "requires," "mandates," or "compels," rather than "prohibits" or "forbids." And that usage of "enjoin" in its "positive" sense in Minnesota court decisions goes back to 1858. "There are many provisions in constitutions and statutes which … are construed to be only directory [as to] the performance of the act enjoined or permitted." Supervisors Ramsey County v. Heenan, 2 Minn. 330 (Minn., 1858) (where applying the label "directory" means that failure to do the enjoined, i.e., directed, act is not fatal to its effect). Nor was that isolated. "Enjoin" in a "positive" sense re-appears (in Minnesota) twice in 1861: "[I]f [a] creditor … omits to do any act … which his duty enjoins him to do, [he may lose certain rights]." Huey v. Pinney, 5 Minn. 246 (Minn., 1861). "[The new system] … permits, even enjoins, [handling all inter-party issues] … in one suit [rather than separate ones]." Morrison v. Lovejoy, 6 Minn. 224 (Minn., 1861). In short, "enjoin" in both senses has long been well attested in the law, as well as in everyday use.
J.W. Brewer said,
May 29, 2019 @ 3:29 pm
Thanks to Michael Trittipo for flagging Lambert v. Blackwell. It is worth noting that then-Judge Chertoff did not cite Grice directly but only as quoted in a then-recently-published piece in the Stanford Law Review written by a legal academic. (I suspect that article in turn may have been drawn to His Honor's attention by an eager law clerk, as Judge Chertoff, like his two then-colleagues who concurred in his opinion, was too sensible a fellow to be randomly browsing the Stanford Law Review on his own.)
And while I was not previously familiar with the academic who wrote that article despite various mutual connections (e.g. google reveals that he has co-authored papers with one of my own former teachers) it turns out that before selling out and going to law school himself Prof. Henry E. Smith had first obtained a Ph.D. in linguistics (from Stanford, in 1992). So on the one hand maybe it's good for someone with that level of training in linguistics to have ended up in the legal academy, but on the other hand it might be better if a higher percentage of lawyers and judges whose academic focus had been in another discipline had gotten just enough of a smattering of exposure to linguistics to know who Grice was.
Andrew Usher said,
May 29, 2019 @ 10:50 pm
The 'Gricean maxims' don't work so well for consumer advertising, as I'm sure many have realised. I'd reckon they were cited not so much because of a consistent philosophy as to find some respectable quote to justify the judges' position, and one of them remembered Grice.
I don't find AB's ad campaign to be our of line at all. First, '100% less corn syrup' can't be comparing zero to zero (and that's obviously not what they intended) because that would be dividing by zero, and we know we can't do that. It is just what Gregory Kusnick said: the corn syrup they use, we don't use (100%).
Adrian:
It might seem strange, but AB has been brewing with rice for a long, long time (both regular as well as light beer) and is probably entitled to be proud of their recipe.
Also, it seems strange for the court to fixate on the 'fact' that no corn syrup appears in the final product. Well, it's altered by fermentation, but it doesn't just disappear! The same is true for barley; is it misleading to say that barley is in beer? In any evern, it looks disturbingly like some judges are trying to protect a large company by censorship of truthful advertising against it, something that's been disturbingly common in the US.
k_over_hbarc at yahoo dot com
Michael Watts said,
May 30, 2019 @ 1:34 am
This is not correct; there's no division involved. 100% less corn syrup than zero is zero, as 100% less than anything is zero, because x – x = 0 for all x. When x is zero, that's still true, in the particular form 0 – 0 = 0. There is a multiplication (applying the percentage to x) and a subtraction (reducing x by whatever percentage of x), but no division.
Michael Watts said,
May 30, 2019 @ 1:38 am
I'll also point out that the argument being litigated was that
This is a comparison of what the beer contains, so yes, it's comparing zero to zero.
But I feel better about my comment above that "most people will resist interpreting the claim that way even if you explicitly tell them that that's what it means".
Andrew Usher said,
May 30, 2019 @ 7:34 am
Agreed with the last, certainly, because of the meaning or 'less' and 'more'. '100%' is a modifier, to be '100% less' you first have to be less, and even in mathematics zero is not less than zero. This is how everyone (including the courts) would interpret it. I was wrong talking about 'dividing by zero', yes – even if you understand percentages as division (I do) 0 would be the numerator, nor the denominator, here.
More to the point is my statement that 'contains' is somewhat fuzzy here. Corn syrup is added to the stuff at some stage; it is not removed as such, but it is changed by fermentation. Would it be correct and non-misleading for MillerCoors to assert their beers 'contain zero corn syrup'? I think it's context-dependent in matters like this.
Haamu said,
May 30, 2019 @ 10:28 am
No, Andrew Usher, you weren't wrong. Division by zero is involved.
There are two different situations, and two different formulas.
Michael Watts, you're correct to this extent: if Bud knows the Coors level (C) and the desired percentage decrease (P), Bud can calculate the value of B (how much syrup they want to end up with). That formula is
As you've noted, P = 100% is a special case where B is always 0, regardless of the value of C.
But imagine if Bud knows C and B empirically and wants to know what claim they can make. That's the relevant circumstance here. In that case, we must solve for P, and we get
So when the baseline C is zero, P is undefined. It actually takes on all values. A 100% reduction takes you from 0 to 0, sure, but so does a 43% reduction or a 50000% increase. That's what makes the claim meaningless.
So a 100% decrease may be a special case, but then a 100% decrease from a baseline of zero is a special case of a special case.
Gregory Kusnick said,
May 30, 2019 @ 2:10 pm
From a practical standpoint, the residual sugar left over after fermentation is never going to be precisely zero. If MillerCoors is claiming it is, they're not being honest. And presumably there are other trace components of corn syrup that do not get consumed during fermentation and that someone with (say) a severe corn allergy might legitimately be concerned about.
Suppose we were talking about milk sugar instead of corn syrup. Could a brewer get away with claiming their milk-sugar beer was vegan? Clearly not.
Michael Watts said,
May 31, 2019 @ 5:31 am
Again, this is not correct:
1. The relevant circumstance here is that Anheuser-Busch made the claim "we have 100% less corn syrup than they do". You can evaluate the truth of this very easily by looking at how much corn syrup Bud Light has; if that amount is zero, the claim is true. But in general, you can evaluate the claim, still very easily, by checking the numbers against the equation B = C – PC. If B is equal to C – PC, the claim is true.
2. The question of "given B and C, what values of P can we claim?" is not particularly relevant, since what's being litigated is the claim they made, not the process by which they chose to make it. However, as you note, all values of P are legitimate answers to this question in the case where B and C are both zero. That fact does not in any way make the claim meaningless for P = 1. It still holds the same meaning as it does in any other context, to wit, that B = 0.
Andrew Usher said,
May 31, 2019 @ 7:54 am
The mathematical quibbling doesn't interest me. Maybe I was, or maybe I wasn't, correct in mentioning 'dividing by zero'. The point in that in real life, comparisons between zero and zero don't work; they're necessarily either wrong or misleading. Even saying 'as much as' (mathematically unquestionable) would imply the existence of some.
I don't think I have any more to say on this matter.
Haamu said,
May 31, 2019 @ 10:18 am
Andrew: "Quibbling"? Well, that's what I get for trying to demonstrate that your intuition has a legitimate basis. I know some people find that sort of thing annoying.
Michael — re this:
and this:
both miss the point that the claim is not just about B, but about the relationship of B to C. So, yes, there's a meaningful and true claim about B, but there's also a meaningful and false implication about C.
And as for "true": the simplest form of the claim, that P = 100%, is not true in a formal logic sense, given that its contrapositive isn't true. (I.e., as we've agreed, it can't be claimed that 100% is the only number that expresses this relationship between B and C.) And yet our intuition is that contrapositives are generally true. That's just a different way of expressing where our erroneous inferences about C are coming from.
So, hopefully, I've demonstrated that I can be just as tiresome about logic as I can about math.
Andrew Usher said,
May 31, 2019 @ 7:14 pm
My frustration was not about that, but about the fact that you were continuing to take the argument in a direction it shouldn't have been going in te first place (since no one is truly arguing in favor of comparing zero to zero).
And then, you come to the conclusion that "there's a meaningful but false implication about C, which is the same thing as I said without needing the math, which as you've shown sets no constraint on C at all.
Michael Watts said,
June 1, 2019 @ 5:49 am
"P = 1" and "B = C – PC" are, from a formal logic perspective… propositions. They are not implications and don't have contrapositives.
Would you try to write out the claim you think is being made, from a formal logic perspective, with arrows and negation signs?