How to nail down clues

Old college threads.
Locked
User avatar
theMoMA
Forums Staff: Administrator
Posts: 5993
Joined: Mon Oct 23, 2006 2:00 am

How to nail down clues

Post by theMoMA »

What makes a good clue? A few things leap to mind: it's dense (avoids superfluous language), interesting (focuses on clues reflecting why the answer is important), clearly written (avoids confusion), factually correct, and specific. Roughly speaking, I've tried to list those criteria from least to most essential. This is a post about specificity, perhaps the most important characteristic of a tossup clue.

To me, "specific" means that each clue in the tossup unambiguously maps to one (and only one) answer line: the answer line on the paper. (I say "one answer line" rather than "one answer" to reflect the fact that tossups often ask about things with multiple names, or a bundle of related ideas with the same name, etc.)

It strikes me that people often do not take as much care as they should to ensure that their clues, especially their leadins, are truly specific to their answer lines, something that Seth Teitler calls "nailing down" the clue. Once you recognize potential specificity pitfalls, it doesn't take very much to nail down each clue so that it points only toward the answer line and nowhere else.

Consider some examples, which are hypothetical, but adapted from questions I recall seeing in recent years:

1. "From a position atop a ridge, the victorious commander in this battle attacked the right flank of the losing side's formation in a single envelopment maneuver."

Perhaps a large enough mass of unspecific details has congealed into a specific whole here, but why chance it?

"From a position atop a ridge, the 21-year-old victorious commander in this battle attacked the right flank of the losing side's formation in a single envelopment maneuver."

This small addition provides an interesting tidbit; the future Prince of Condé was extremely young when he won at Rocroi. It is very unlikely that there are any other battles fitting these clues. You could also name a geographic formation on the battlefield, provide the name of one of the minor commanders or soldiers who participated in the flanking maneuver, etc.

2. A man with this characteristic in Ontario's Legislative Assembly, Gary Malkowski, introduced a 1993 Ontarians with Disabilities bill.

The problem with this formulation is that, by the plain words of the question, every characteristic of Gary Malkowski is the correct answer.

The first man with this characteristic to serve in Ontario's Legislative Assembly, Gary Malkowski, was an NDP member of parliament who introduced a 1993 Ontarians with Disabilities bill.

This one has a relatively easy fix; there is only one first deaf man to serve in that esteemed body, and more importantly, Malkowski wasn't the first man of any other sort to serve. But it's important to remember that clues based on a characteristic of a person or thing need to reference the answer specifically enough so that there aren't multiple correct answers. This is especially true for leadins. Be wary of the construction "of this type" in the leadin, and make sure you're not relying merely on heuristic for people to know what "type" you're looking for.

3. The staple crop of these people was amaranth, which was often eaten with honey.

I see this kind of problem a lot when editing social history questions. The issue here is that, although this clue does apply to the Aztecs, it also applies to several other groups around the area.

Diego Duran wrote that the staple crop of these people was amaranth, which was often eaten with honey.

Naming a specific source is a good way to get around issues like this.

4. Fighting forces such as [X] and [Y] were active during this period.

This is a generalized example of a classic kind of "unspecific" clue that involves simply listing participants in a period or war or movement. A problem arises when one or more of the participants overlapped with other periods or wars, etc.

During this period, the fighting force [X] participated in the Battle of [Z].

This is also extremely easy to get around. Instead of lazily listing participants, pick one or two and provide a clue about something specific that they did.

5. This modern-day country was home to the historical Zagwe Dynasty.

Historical kingdoms are good ways to test about the ancient history of modern-day countries. Unfortunately, several of them actually spanned multiple modern-day countries; the Zagwe Dynasty, for example, likely controlled modern Eritrea.

This modern-day country's city of Lalibela was the capital of the historical Zagwe Dynasty.

Picking a specific feature of the historical kingdom that was, in fact, found in the modern-day country that you're writing about is an easy way to get around this problem.

All of these above scenarios (and many more I'm sure you could devise) are relatively easy to get around if you're looking out for them. And that's the purpose of this post: to get you looking.
Andrew Hart
Minnesota alum
User avatar
Mike Bentley
Sin
Posts: 6461
Joined: Fri Mar 31, 2006 11:03 pm
Location: Bellevue, WA
Contact:

Re: How to nail down clues

Post by Mike Bentley »

3. The staple crop of these people was amaranth, which was often eaten with honey.

I see this kind of problem a lot when editing social history questions. The issue here is that, although this clue does apply to the Aztecs, it also applies to several other groups around the area.

Diego Duran wrote that the staple crop of these people was amaranth, which was often eaten with honey.

Naming a specific source is a good way to get around issues like this.
This can work sometimes, but only if your source is famous enough for people to reasonably have a chance of knowing. It doesn't really help anyone if your ambiguous clue that applies to lots of different answer lines attributes something to Joe Blow from the 3rd page on a Google Books search.
Mike Bentley
Treasurer, Partnership for Academic Competition Excellence
Adviser, Quizbowl Team at University of Washington
University of Maryland, Class of 2008
User avatar
Skepticism and Animal Feed
Auron
Posts: 3238
Joined: Sat Oct 30, 2004 11:47 pm
Location: Arlington, VA

Re: How to nail down clues

Post by Skepticism and Animal Feed »

What about prior clues that eliminate other answers.
Bruce
Harvard '10 / UChicago '07 / Roycemore School '04
ACF Member emeritus
My guide to using Wikipedia as a question source
User avatar
Cheynem
Sin
Posts: 7219
Joined: Tue May 11, 2004 11:19 am
Location: Grand Rapids, Michigan

Re: How to nail down clues

Post by Cheynem »

The challenge I think (as Andrew knows full well) is that it is easily possible to have a "nailed down" clue that is equally as unhelpful as something not nailed down. Thus, while the addition of many of these phrases (21-year-old, first to become, ___ writes, etc.) makes it specific, it might not make it any more buzzable. The question writer's job, within reason and however appropriate, is to do both. I agree that the addition of many of these phrases are ways to resolve the issue in which "it's pretty obvious what I'm looking for and it's a good clue but I need to make it specific," which happens a bit.
Mike Cheyne
Formerly U of Minnesota

"You killed HSAPQ"--Matt Bollinger
User avatar
Mewto55555
Tidus
Posts: 709
Joined: Sat Mar 13, 2010 9:27 pm

Re: How to nail down clues

Post by Mewto55555 »

Another (slightly controversial) way of doing this: using good fucking clues.

EDIT: To be slightly less snarky, I think this is a problem that arises a lot in clues that are generally pretty dumb. In my experience, specificity most needs to be added when a clue clearly refers to a very small number of things in an obviously-problematic way (e.g. doubly-eponymous science things), where it can easily be dealt with by adding the brilliant "alphabetically-prior" to the question. If amaranth is a truly important thing about this culture, then there's probably a better way to clue it without name-dropping a random dude who doesn't help people buzz; if Diego Duran is truly an important dude, but amaranth is somewhat ubiquitous, then the clue should be about him instead!
Max
formerly of Ladue, Chicago
User avatar
Ike
Auron
Posts: 1063
Joined: Sat Jul 26, 2008 5:01 pm

Re: How to nail down clues

Post by Ike »

I'm 100% behind Andrew's post.

I actually want to take this thread and say the following about confusion and interesting clues: if you are writing a tossup so that it's only understandable by people who know the clue, you're doing it wrong. I myself made this mistake numerous times, and I want to post the example that helped in getting me to fix this aspect ofmy writing. I once wrote a tossup on _translating Dante_ for NAQT:
Charles Eliot Norton's use of the third person in this task resulted in a prose narrative. Clive James used quatrains for this task in a 2013 publication. Barbara Reynolds completed this task for Dorothy Sayers, who died before she finished. The first American to attempt it was {Henry (*) Wadsworth Longfellow}, who began with the lines "Midway upon the journey of our life I found myself within a forest dark." For 10 points--name this task of rendering a poem by Dante into English.
To which Andrew Yaphe, the tossup's editor, wrote back to me:
On the whole, I like this question; I think the clue order is good, the answer choice is creative yet gettable, and the selection of clues is solid (nice to see Clive James's recent version getting worked into a question). I'm only sending this back because the first sentence is, to my mind, confusing. I understand--sitting here and staring at the text of the question--what you're trying to say: i.e., that Norton produced a translation of the Comedy in prose (rather than verse), and that his prose version is written from the third-person perspective (rather than the first-person of the poem itself). But I can only unpack that because I'm sitting here reviewing the sentence at my leisure. Hearing this in a game, I think, I would only be baffled by "used the third person." (I would want to say "used which third person? used him to do what?") (Though I might just buzz in blindly on Norton, knowing him to be part of the circle of 19th-century Boston Dante enthusiasts; but if you don't know who Norton is--and we are presuming most players don't, or we wouldn't make this the lead-in--I think you will merely find this perplexing.)

The challenge, then, is to communicate this information in a way that (a) won't prematurely tip off the answer but (b) will still be comprehensible. I'm not quite sure what the best way to do that would be. "This task was accomplished in the 1880s [or whenever] by Charles Eliot Norton, who used third-person prose for his version"? That's not perfect, but something along those lines would, I think, be an improvement--see if you can do better
It's really important to write tossups and bonuses so that players find the material interesting and understandable. If you're just ramming down decontextualized clues or clues that are only interpretable by people who know the subject at hand, you're making the game incredibly dull for most people. Obviously no one is going to be perfect, and in some categories this is a bigger challenge than in others (science!).
Ike
UIUC 13
User avatar
The King's Flight to the Scots
Auron
Posts: 1645
Joined: Mon Jan 26, 2009 11:11 pm

Re: How to nail down clues

Post by The King's Flight to the Scots »

Mewto55555 wrote:Another (slightly controversial) way of doing this: using good fucking clues.

EDIT: To be slightly less snarky, I think this is a problem that arises a lot in clues that are generally pretty dumb. In my experience, specificity most needs to be added when a clue clearly refers to a very small number of things in an obviously-problematic way (e.g. doubly-eponymous science things), where it can easily be dealt with by adding the brilliant "alphabetically-prior" to the question. If amaranth is a truly important thing about this culture, then there's probably a better way to clue it without name-dropping a random dude who doesn't help people buzz; if Diego Duran is truly an important dude, but amaranth is somewhat ubiquitous, then the clue should be about him instead!
Yeah, I side with Max on this one. I don't think "21-year-old" will help many more players, in this case. As a general rule, it's tough for players to combine multiple non-unique sources of information into a useful clue, even if in theory their combination is uniquely identifying. In this case, "21-year-old" + "right flank" + "enveloping maneuver" may be unique in principle, but because the human brain is not SQL searchable, it doesn't do much good at game speed.

I think these types of clues should only be used after you've contextualized the answerline. The most useful forms of context tend to be geographic, temporal, and thematic. If the question had already established that the battle took place in northern France, or in the 17th century, or within the broader narrative of the Thirty Years' War, those clues start to make a lot more sense. That's because when you're asking a player to search across multiple criteria on the fly, it's important to limit the dataset. It's just impossible to check every battle from Cynoscephalae to Sinope.

Properly using contextual clues is the hardest part of writing history questions, because you have to judge how far you can narrow down the geographic and temporal circumstances of your answerline without giving it away. It's also essential if you're going to try to use the types of technical clues that Andrew* is describing.

*And I should add that Andrew is very good at this type of player empathy in his writing, and my additions to this thread should not be taken as an attack on him or his ideas.
Matt Bollinger
UVA '14, UVA '15
User avatar
theMoMA
Forums Staff: Administrator
Posts: 5993
Joined: Mon Oct 23, 2006 2:00 am

Re: How to nail down clues

Post by theMoMA »

I probably didn't convey this very well, but my intention with this post was to discuss how to nail down a clue that you've already determined is good (i.e. dense, interesting, clearly written, and factually correct). I used "interesting" to mean that the clue "reflects why the answer is important," so in that sense, I could just as well have used "famous enough for people to reasonably have a chance of knowing" (as Mike B. says); "buzzable" or "helpful" (as Mike C. says); "good fucking clues" (as Max says, and as he elaborates on the term); or "interesting and understandable" (as Ike says). If my illustrative examples don't convey this fully, well, that's probably because they were less-than-lovingly curated, and for that, I apologize. But my post was, if opaquely, proceeding as though we presupposed that the clue fit these criteria; I completely agree that nailing down an otherwise deficient clue doesn't rescue it (although it does make it less deficient by resolving the specificity issue).

It strikes me that an aspect of the "interesting" criterion that you guys have pointed out (which I didn't initially articulate) is that the additional information that nails down the clue should itself be interesting (or whatever you want to call it). In other words, the specificity can't simply be sourced to a random PubMed article or a third-tier Google Books result. The specificity has to be functional; people have to actually be able to distinguish based on it. This is a good point that I'm glad you guys have drawn out.

Both Ike and Matt have a very well-founded point about context. I don't know much about tactics, but I wanted to make an example of a tactical clue, because they're often written so that their specificity is suspect. I happened upon the fact that Rocroi appears to be a "textbook" flanking/single envelopment battle. Now, does that mean that I should immediately hone in on this clue as the leadin? As Matt says, probably not. Asking a player to piece together several disparate pieces of information, divorced from all context, is almost always unrealistic. But after establishing some geographic or temporal context with other good, specific clues, a clue along the lines of "The winning commander of this battle, who executed a textbook single envelopment from a position atop a ridge, was a 21-year-old duke who [some Grand Conde clues here or whatever]." To me, this is a good use of a tactical clue, because it ties the tactics to why they're important (they were successful in a textbook fashion, and they were an early success for a renowned tactician).

Breitenfeld is apparently also a textbook single envelopment, which illustrates the need for clues that are functionally specific. This seems relevant to Bruce's question about answers ruled out by previous clues. It would be particularly frustrating to realize that a tossup was talking about a Thirty Years War battle and neg with Breitenfeld on a clue whose entirety is something like "this battle's victors executed a single envelopment of the enemy's left flank." (This is true of both battles, it appears.) Even if you just nailed it down in the most basic way ("this battle's 21-year-old winning commander executed a single envelopment of the enemy's left flank"), you take away reasonable complaint; Gustavus Adolphus wasn't notably young at Breitenfeld.

I think it's still safe to say, as Matt might, that clues should take care to distinguish on things that are less fine-grained than whether you can parse Gustavus Adolphus's age at game speed. But (a) attention must be paid, and (b) again, this is just the most basic way to nail something down, and it still mostly gets the job done. If you're on the lookout for the loose threads of unspecific clues and are willing to weave them into the fabric of the tossup in a more contextually rich way (such as my example two paragraphs above), you can obviously do an even better job.
Andrew Hart
Minnesota alum
Locked