Cody wrote:Often, tossups will instruct moderators to prompt on supersets that apply for nearly all, but not all, clues. This is okay in my book (I've done it). The key distinction is that the promptable answer does apply to many clues in the tossup (perhaps "all" was a bit too hyperbolic). There are even situations where prompts due to a specific clue are deserved, as I'm sure I've done.
Matt Weiner wrote:But expecting to get credit for a WRONG ANSWER because "this one clue applied to it and I didn't know the other clues" [...] is not workable in the protest system.
This (hypothetical Haydn, etc example) is just the Rhode Island thing all over again. You have to say the right answer. Right answers are right and wrong answers are wrong. "The question was bad" is not grounds for protest. "I didn't know the leadin" is not grounds for protest. "I want to buzz on the eighth clue of a question and ignore all the clues surrounding it" is not grounds for protest. Any question that properly specifies a correct answer must be played as written [...]
While this is certainly a reasonable position that one could argue for, in the abstract, it does not reflect the actual logic behind how questions are being written and how prompts are being decided in the majority of tournaments each year. Even people as rhetorically hard-line as Matt Weiner do not usually enact their own stances in their editing. Here are the first four sentences from a tossup from last year's History Bowl on the Beirut Barracks Bombing:
A month prior to this event, the May 17 Agreement was cancelled. A few months after this event, an American named Malcolm Kerr was murdered at a nearby university. One justification for this event was the gunfire support offered by the Arthur W. Radford and the John Rodgers. In the aftermath of this event, the U.S.S. New Jersey heavily shelled the Bekaa Valley.
This has a prompt instruction throughout for "Lebanese Civil War", even though an answer of "Lebanese Civil War" is incorrect for every single one of these clues
, because the chronological markers "prior to this event", "after this event", "justification for this event", and "aftermath" would be completely wrong in relation to the war as a whole (all of these things took place in the middle of the war, rather than before or after it). Here, the prompt is purely motivated by the answer being a subset of a larger conflict, rather than because the majority of the clues are also correct for the superset answer being prompted.
This is of course in no way invalidates Cody and Matt's arguments that their standards (that nearly every clue must correctly apply to the prompted thing in order for a prompt to be justified) should be
used. But it certainly erodes Matt Weiner's historical/naturalistic attitude in all of his posts about protests that his stance is just the way quizbowl works and always has; in fact, quizbowl rarely works like this in reality. That Weiner's stance is merely rhetorical and is not supported by practice (even his own) can be confirmed by going through any of the major tournaments of the past few years to see when prompt instructions are given, for what answers, and why. In order for Weiner's stringent protest requirements to be practical, writing practice would need to be reformed quite a bit to conform to them.
theMoMA wrote:I'd basically say that there are two types of prompts: permissive and mandatory ones. A mandatory prompt would be like, prompting on "Johnson" when you require "L. Johnson" or "LBJ." A permissive one would be an editor or writer deciding that, even though these things aren't perfectly nested within each other, there is a justification for a generous prompt (such as the two examples John offered above, where certain connecting language would technically rule an answer out, but a prompt still seems entirely justified).
I would argue that "should have been prompted" protests should be upheld, and a correct answer awarded outright, when the missing prompt was mandatory. I would argue that "should have been prompted" protests should be outright denied if the prompt was permissive, since it was by definition a judgment call, and the judgment rested with the writers or editors (or moderators), who elected not to prompt.
I disagree with this dichotomous distinction, as Andrew has stated it, because it omits a crucial fact of the Haydn case: the second clue (about the Drumroll symphony) does not in any way distinguish between the superset (Haydn's Symphonies) and the subset that is its actual answer (London Symphonies). Someone buzzing on the second clue, who does not know the first clue has absolutely no way
to know which answer is required.
The problem I have with what Andrew says here and with what Cody says earlier--when he suggests that protests should inherently carry a more stringent standard than prompts--is that it makes it impossible to know what to do when confronted by a clue like the second line of the Haydn tossup. Let us say that I know exactly what superset and subset it belongs to, but simply do not know the previous clue. What should I do? Do I buzz now, and say the superset, hoping to get prompted because most editors do that, but knowing that I will get an unprotestable neg if the editor happened not to include a prompt instruction (since there are no rules obliging him to do so)? Or do I wait another sentence, in the hopes that the next sentence will clarify which set is being sought, risking that my opponent knows the next clue and will beat me, even though he has less knowledge. The necessity in engaging in this kind of thinking--in which I need to guess the editors' whims or (even worse) rely on my knowledge of the editors' stance on this ideological question (from having read these forum posts) to guide my gameplay--is a basic characteristic of "bad quizbowl". Rules reforms need to be enacted to eliminate these situations.
What exactly should we do about them? To my mind, there are only three sensible positions:
(1) All common-links need to be written in such a way that every single sentence distinguishes the subset from its superset, so that you never need to rely on prior clues to know which the question is going for, provided you sufficiently understand the individual clue that you buzzed in on. This is the only situation in which Matt Weiner's approach makes sense.
(2) Subsets that are correct for the clue that you buzzed in on should always be either acceptable or anti-promptable, so that players can default to saying the subset that the clue they buzz in on indicates, and can retreat to the superset once they are anti-prompted.
(3) Any clue that fails to distinguish between superset and a subset, when the answer is the subset, should have a prompt instruction on the superset.
No one currently writes in style (1), and it would be impractical to implement, because research practices in question writing are simply not stringent enough for it to be realistic. Also, it would mean that (e.g.) a clue on Haydn's Drumroll Symphony in a London Symphonies tossup would always have to say "the eleventh of these", thus rather giving the game away. Option (2) has been tried in mACF formats (and quite common in NAQT, where components of common-links are usually acceptable), and I'm willing to support it, but people apparently are not very fond of it in mACF formats, and have decided that they don't want it to be part of standard quizbowl practice. This leaves Option (3), which I strongly advocate should become part of standard quizbowl practice.
DumbJaques wrote: There is absolutely no need to have a rule that permissive prompt protests should be outright denied. That is, if I'm writing or editing, and I just forget to put in a permissive prompt (or more likely, there's a clue that absolutely WOULD have caused me to add such a prompt, but I just didn't know about that aspect of it). In this case, I hope someone does protest and allow me to address the issue. That is to say: Why should there be a sort of moral luck to this rule? If the same writer forgets to add a permissive prompt in one question but not the other, why should some players suffer and others not?
This. Weiner and I had a 40-minute argument about this point in a rules reform IRC meeting last spring. Weiner's position seemed to be that questions have to be understood as reflecting the final intentions of the editor; therefore a lack of prompt instruction has to be treated as a refusal on the editor's part to prompt. Of course, as Chris has just pointed out, this is a practically-motivated fiction: editors frequently omit prompt instructions through oversight. My argument was that this fiction probably needs to be adhered to in cases where protests are being adjudicated by TD's rather than editors; the TD has to take the question as the editor's word on the issue. However, there is absolutely no reason to adhere to this fiction when the editor himself is the one in charge of adjudicating protests (e.g. as is usually the case at ACF Nationals), and to prevent him from being able to correct his oversights. I suggested that a clause be added to allow editors to review these protests, in those specific cases. After all, such protest resolutions are completely painless: all the editor needs to do is say "Whoops! I totally meant to prompt there!" or "Nope. I deliberately did not prompt there. The neg stands".
At the meeting, I had the support of the majority by the end of our argument, when this issue was voted on, and I was told that as a result of the vote, this suggestion would be adopted for last years' Nationals. I don't see it in the current version of the ACF rules, and I'm not sure what happened there. If this needs to go to a vote again, then I look forward to bringing it up at this year's ACF meeting.