This article proposes that we explicitly classify a type of theory shell that has proven more often than not frivolous and exhausting. My hopes are that, with this classification, it will be easier to identify frivolous theory and the community will more wholly be able to engage in norm-setting practices that help curb the use of these shells.
I made up this category as a high school senior when I was working on figuring out the intricacies of theory debate (jury is out on if I ever succeeded or if I was even original in this identification). LOS stands for Link of Omission Shell. I was happily taking advantage of the practices prevalent in Texas circa 2016-17, when theory debates were more the norm than the exception, and all of my 1ncs had some kind of bidirectional shell that could be thrown out there. The concept of bidirectionality started to interest me, and as I mulled it over I came to the realization that one side of the two possible shells was much, much stronger than the other. Bidirectionality, as I understand it, can best be explained as an action that can be argued on either side by debaters proposing a norm. The best example of this is what we’ve called “spec” shells. You will always, always see debaters prepared for debates with one of two options at the ready; “you must specify x” or “you must not specify x.” There is a sufficient argument on either side such that the shell can generate viable offense and prove a readily available generic. It is my belief, however, that the “you must” side of this generic should start to go away with the advent of looser thresholds on reasonability, because those shells rely upon a Link of Omission.
Link of Omission Shells are shells that attempt to obligate a debater to do an action rather than to discourage an action that has already occurred. This is easiest demonstrated by spec; “You must provide a definition,” “You must provide an actor,” You must provide the status of the counterplan in the speech.” All of these ask for something retroactively from the debater, with the punishment for not having done the action in question being a loss. I believe that this is wholly disproportionate most always, and that the debate community should on balance discourage these shells.
First and most importantly, these shells create an incentive to move towards the most minimal requests that can justify the disproportionate response of a loss. This invites an infinite number of arbitrary shells, whose offense boils down to “you would have made the debate slightly easier for me.” How much offense can genuinely be created on the argument “you must provide status in the speech?” Realistically, none (or very, very little) that’s altogether convincing. What is readily available is defense, which competing interpretations limits from being able to win the debates. This means that debaters can throw these weak shells in and comfortably spit out “competing interps > reasonability” with a ton of work on risk of offense being sufficient, because there’s nothing to weigh against the initial complaint. Having a more coherent explanation for why judges should grant leniency to the respondent in these cases will give judges a much stronger justification for why they pulled the trigger on reasonability. My vision is debaters responding with something like “Prefer reasonability: This is an LOS, which invites marginal abuse claims with little response and limits me to defensive answers, which is not an accessible strategy under competing interps.” Arguments of this flavor let judges lean towards the side that is not meeting the burden in question and do so with a bit more communal weight behind the warranting.
Second, there is no way to retroactively resolve the issue in question. Drop the argument grew as a way to shift out of the abuse in question, giving debaters the opportunity to forego the theory debate, resolve the abuse, and continue with substance. This option has totally disappeared when you haven’t done anything that you can drop, which takes the limited responses and gives one fewer paradigmatic response.
Third, which is strongly related to my first argument, the punishment to these shells is nearly nonexistent. Because of how difficult it can be to generate offense, RVIs are largely unavailable as they don’t pair especially well with a defensive approach to the claim. This makes the risk factor very low for the reader, opening the door to plenty of frivolity.
The reason that the “must not do x” shells are better is that debaters can defend the practices they took on. There’s a limited number of those that exist in the round that can be called out, they can employ drop the argument, and they should be able to offensively assert that the practices they took on were viable practices. It is much, much harder to defend yourself for not having done things that were never explicitly required of you. Think about this in a real world context; how often do you get punished for things that you didn’t do that you could have done? The law and general society is far more responsive to existing harm than it is about using punishment to encourage improvements, and I don’t see why debate should not follow that logic.
There may be a few responses to this argument. The first is that these shells rightfully call out abusive omissions by the respondent. The best example of this is that disclosure falls into the category of LOS, and there are very, very compelling arguments for why we should enforce losses against debaters who don’t disclose. Fair argument, and I think that if you are so convinced that certain omissions are harmful, you should read the shell and be willing to defend against the reasonability argument I propose. A debater winning that we should use the paradigm of reasonability does not mean they have proven themselves reasonable – you should still be able to prove that their omission was so damaging that the judge can’t rule them reasonable. Disclosure is a clear example of the abuse story being able to beat back the defense that’s available, whereas actor spec is a clear example where CX checks is so dominating that this will be harder for the theory reader. My goal here isn’t to say LOS is a non-option, but that the standard for these should be much higher than it is.
The only other argument that I can think of is that this puts way too much power into the hands of a judge, instead of letting the debate play out. This argument is more about the reasonability paradigm itself, so I don’t think it’s on me to justify this categorization against it. I will say, however, that theory debaters should remember that the ultimate aim of theory is not having more theory, but less. We should be able to hone practices enough for theory to be a nonexistent practice in the long run, and norm-setting as a side constraint means that there is a certain element of communal input here that lets the judge assert themselves where they otherwise should not.
I am hoping that this article helps debaters compete against theory shells that can be frustrating to respond to!