High School/College Partnerships

Communications Studies Supplements

The Disposition of Counterplans and Permutations:
The Case for Logical, Limited Conditionality

Roger Solt

2003 - Mental Health Policies: Escape from Bedlam?

The conditionality of counterplans has been one of the most persistent controversies in debate theory. Reconnoitering the dim archives of my long-term memory, I can vaguely recall counterplan conditionality being a theoretical issue in the early 1970s. This was in an era in which counterplans really were a "theoretical" issue because almost no one ran them-they were considered a "strategy of desperation." When the age of the counterplan did finally arrive (starting in the mid-1970s and really gaining force in the early 1980s), the debate over conditionality became both more practically relevant and more heated. For a time it was THE central theoretical issue in academic debate (stirring some of the same sound and fury that now surrounds the kritik) since it was the question which most clearly divided the two leading paradigms of the day, policy making and hypothesis testing. (Policy makers, with some notable exceptions, were against conditionality; hypothesis testers favored it.) As debate theory and practice have continued to evolve, the conditionality controversy has become less central.

Still, arguments concerning conditionality occur in a high percentage of the debates in which counterplans are run. And the conditionality debate has become more muddled, in part because it has become separated from more overarching paradigms of debate theory, and in part because of the emergence of that strange form of quasi-conditionality referred to, rather inelegantly, as "dispositionality." Not only is "dispositionality" an ungainly word; it is also an ambiguous word, inconsistently employed. In innumerable debates the following exchange has occurred: "What is the status of the counterplan?" "It's dispositional." "What does that mean?"

This essay is written in part to answer that question-to cope with the reigning terminological confusion-by sorting out the assorted meanings of conditionality and dispositionality. In doing so, I will describe the two traditional, polar positions on the matter: classic, unconstrained conditionality (CUC) and strict, single option advocacy (SSOA). I will argue that the two main theories of counterplan dispositionality fall on a continuum of permissiveness between these two poles and that both versions of dispositionality can be described as forms of logical, limited conditionality (LLC). I will also argue that both forms of logical, limited conditionality are superior to the more extreme alternatives of classic, unconstrained conditionality and strict, single option advocacy.

The Discourse of Dispositionality: An Analysis of Terms

Debate has presumably borrowed the term "conditional" from formal logic. In logic, a conditional statement is basically an "if/then" statement. IF a certain antecedent condition exists, THEN a certain consequence follows. Debaters also frequently employ an "even if" logic. This is commonly seen in weighing arguments, as in "there is no link to the disadvantage, but EVEN IF there is one, the case outweighs. There is also a certain conditional logic involved in dilemmas and double binds, for example, "the plan does not solve, but if it does, then it links to a disadvantage." It has even been suggested that there is a kind of conditionality between topicality arguments and arguments concerning policy desirability. For example, "the plan is not topical, and even if it is, it's a bad idea." (This same argument, however, could be phrased non-conditionally: "the plan is not topical, AND it is a bad idea.")

It should be clear that these kinds of "if/then," "even if," and "either/or" arguments are perfectly common in debate (as well as in other argument settings) and that for the most part their legitimacy is not particularly controversial. What is controversial within academic debate are CONDITIONAL POLICY ALTERNATIVES. The negative team is presenting conditional policy alternatives when it offers more than one policy option to the judge. The extreme example of this occurs when the negative offers two or more counterplans in the same round. What is more common today, however, is for the negative to say that the judge can vote either for the counterplan or the status quo. For example, the negative side might argue that public health services for mental health care should not be increased at all, but if they should be, the states rather than the federal government should increase them.

The affirmative also sometimes offers conditional policy alternatives, that is, more than one policy option, either of which is said to justify an affirmative ballot. The most radical form of affirmative conditionality is probably the theory of alternative justification. According to this theory, the affirmative can present two or more distinct plans in the same round, and if any of its plans is determined to be desirable at the end of the round, they win.

The idea of plan severance is a second form of affirmative conditionality. Assume that the affirmative plan has more than one provision, and that one part, but not every part of the plan is proven to be undesirable. The affirmative might then argue that the judge can reject (sever out) the undesirable parts of the plan and still vote affirmative for the favorable portions, which remain.

A third form of affirmative conditionality is an intrinsicness argument. Here the affirmative argues that a disadvantage is not an INTRINSIC reason to reject its plan (or the resolution) because an additional policy element could be added to the plan that would stop the disadvantage. (For example, Congress could adopt the plan AND ban national missile defense, if that is the disadvantage.) One could imagine intrinsicness arguments being made in a non-conditional way; that is, the affirmative would add a new plan element in the 2AC, but it would then be stuck with it and liable to any disadvantages it entailed. In general, though, intrinsicness arguments have been advanced conditionally. If the intrinsicness argument/minor repair was proven undesirable, it simply dropped out of the round.

A fourth form of affirmative policy conditionality involves permutations, if they are conceived of as policy options the judge can endorse at the end of the round. Permutations have not traditionally been thought of as being conditional, but if the judge has the option of voting for the original plan, OR permutation one, OR permutation two…etc., then this seems just as conditional as saying the judge can vote for the status quo, OR counterplan one, OR counterplan two…etc.

Contemporary debate theory rejects affirmative policy conditionality in most, if not all, of its forms, and with good reason. The theory of alternative justification, first advanced in the 1970s, was never widely accepted. One objection is that it would make debate too superficial. It is hard enough to have an in depth discussion of one affirmative plan within the less than two hours allocated for academic debates. To consider two or more affirmative plans in that time period invites superficiality. This is especially so since each affirmative plan might be answered by multiple negative counterplans, and each counterplan might invoke multiple affirmative permutations. This is argument Malthus with a vengeance. Second, alternative justification encourages argument avoidance. Wherever the negative strongly attacks, the affirmative will be encouraged to retreat, ultimately falling back to whatever policy option the negative was least able to engage.

Third, alternative justification is especially problematic on bidirectional topics. If plan one moves in one direction and plan two in the opposite, then negative arguments which defeat the first plan are likely to support the second. Fourth, alternative justification would let the affirmative run new and/or marginally topical plans with relative impunity. The normal disincentives to doing so are vastly reduced if the new plan can simply be dropped and one can go for the old. In sum, alternative justification seems problematic both on grounds of education and fairness. By making debates highly superficial and discouraging direct clash, significant educational values figured to be lost. And alternative justification also seemed likely to overwhelmingly bias competitive results towards the affirmative. Finally, there is what might be considered a definitional argument against alternative justification: the plan is the central focus of the debate, so the plan should remain a fixed and unmoving point of reference throughout the round.

Most of these arguments also apply to plan severance. (The difference between three plans and a plan with three detachable elements is not very great.) The severance approach also suggests the decoy strategy: put some especially bad provisions into your plan, let the negative concentrate its attack there, and then sever those parts out. No one, presumably, thinks this would make for good debate.

Intrinsicness arguments seem less fatally flawed than either alternative justification or severance. If decision makers can foresee disadvantages to a policy and also see how to avoid them, consistent with the policy's central thrust, then rationally they should do so. And especially if one sees debate as a search for the best policy, then intrinsicness arguments make sense as part of the optimum policy package. Still, the objections, both practical and theoretical, to intrinsicness seem on balance to be compelling.

First, intrinsicness arguments make the plan a moving target, violating the principle of fixed plan focus. Further, intrinsicness arguments can have the effect of changing the plan more than once; indeed, they can make it into a veritable amoeba, sprouting and retracting policy tentacles from speech to speech. The 2AC adds the minor repair/intrinsicness argument to the plan; the negative proves the minor repair disadvantageous; the affirmative drops the minor repair and reverts to its original plan. Or, in what is termed the infinite regress problem, the 1AR offers a new minor repair to answer the disadvantage to the first minor repair, and the 2AR potentially a third minor repair to answer any disadvantage to the second minor repair.

Second, as a practical matter, it is very difficult to find intrinsic disadvantages. Politics DAs, for example, are almost never intrinsic. Third, intrinsicness arguments may be too utopian; that is, they let the affirmative create a world in which its plan is a good idea. Fourth, they are generally non-topical. And fifth, they may demonstrate the insufficiency of the resolution, that is, illustrate that the resolution isn't a good idea unless coupled with other, non-topical elements. These are the most common arguments against intrinsicness, listed largely in descending order of persuasiveness. None of them seems totally decisive, but taken together they probably justify the conventional wisdom that intrinsicness arguments are, on balance, illegitimate in academic debate. The issue of permutation conditionality will be reserved until later.

While affirmative conditionality has few defenders, negative conditionality has many, and it is the various forms of negative conditionality which I now wish to describe. The traditional theory of negative conditionality, associated with hypothesis testing and even defended by some policy makers, is what I have termed classic, unconstrained conditionality (CUC). For hypothesis testers, conditional counterplans merely functioned as inherency (or intrinsicness) arguments against affirmative advantages; that is, they proved that the resolution isn't necessary in order to achieve the affirmative advantage. (To achieve the advantage, we need not do A. We could just as well do B, C, D, or E.) Since counterplans are just logical tests of inherency, there was no real theoretical limit on how many could be introduced.

For traditional policy makers, the plan, not the resolution, was the focus of the debate, but the negative arguably should be able to attack the plan from multiple vantage points. Rational policy makers should consider multiple policy options, and if any one is superior to the plan, then the plan should be rejected. Again, within this theory, there seemed to be no logical limit to the number of counterplans which the negative could advance.

The theory of classic unconstrained conditionality stands in opposition to the perspective I have labeled strict, single option advocacy. Within the SSOA framework, each team is required to advocate one and only one policy option and to do so consistently throughout the round. Thus, in this world, the affirmative can have only one plan, the negative only one counterplan, and if the negative advances a counterplan, the status quo is no longer an option which the negative can also defend or the judge endorse.

Theories of dispositionality, I have suggested, fall between the extremes of classic, unconstrained conditionality and strict, single option advocacy. They entail a position that I have labeled as LLC-logical, limited conditionality. To understand the concept of dispositionality, a little history lesson seems advisable. I have suggested that "dispositionality" is an unfortunate term which debate would be better off without, both for reasons of clarity and of felicity of expression. However, now is also probably a good time to offer a confession: I think that I am (indirectly) responsible for the introduction of the terms "dispositional" and "dispositionality" into debate's lexicon. Perhaps there are other stories which could be told about these terms' etymology, but this is my version of the matter.

In the mid-1980s, I became interested in developing a theoretical middle ground between classic unconstrained conditionality and strict, single option advocacy. In particular, it seemed to me that even if the negative were limited to only one counterplan, there might be circumstances under which the status quo should remain an option available to the judge at the end of the round. I therefore encouraged my teams to advance theoretical arguments in the 1NC concerning the "disposition" of the counterplan. The first article to discuss "the disposition of counterplans" was produced by my former student (and current business partner) Ouita Papka Michel, writing in the 1986 DEBATERS' RESEARCH GUIDE. By counterplan "disposition" we simply meant "what happens" to the counterplan under a certain set of circumstances. One disposition of the counterplan is that if it is proven undesirable OR theoretically illegitimate, the judge still has the option of endorsing the status quo. Or, to put it another way, the negative can always revert to the status quo as its position at the end of the round.

A second disposition of the counterplan is that if the counterplan is shown to be theoretically illegitimate, the negative can revert to the status quo, but if the affirmative only attacks the counterplan's desirability as a policy (that is, if they "straight turn" the counterplan), then the negative is stuck with it. A third disposition of the counterplan is that if the counterplan loses either on theoretical acceptability or on policy desirability, the negative loses. The counterplan in this view is unconditionally non-conditional. This is basically the strict, single option advocacy position. The negative team has only one policy advocacy. In a counterplan round, that is the counterplan. And if the counterplan is defeated, the negative loses (unless they win topicality). There is literally nothing to vote negative for.

Some time after the term "counterplan disposition" emerged out of the matrix of Kentucky debate, the terms "dispositional" and "dispositionality" became common, paralleling the terms "conditional" and "conditionality." I regard this terminological evolution as unfortunate. My objection is in part esthetic. "Conditional" and even "conditionality" are at least words in the English language employed beyond debate. "Dispositional" and "dispositionality" strike me as ugly debate neologisms which we could live very well without. But more importantly, to simply say that a counterplan is "dispositional" literally means nothing. For it to be dispositional means that something happens to it under certain conditions, but that term, standing alone, neither specifies what happens nor the conditions under which a given disposition of the counterplan takes place. There are several different dispositions possible for defeated counterplans, and unless you specify which one of them you intend, the counterplan's disposition remains indeterminate.

Between the extremes of classic, unconstrained conditionality and strict, single option advocacy, I have suggested that there are two main options for counterplan disposition. Following the felicitous suggestion of Steve Mancuso, I will refer to these options as "negative counterplan disposition" and "affirmative counterplan disposition." The theory of negative counterplan disposition is that the negative, like the affirmative, gets one and only one proposal for change but that the status quo always remains a logical option available to the judge. The theory of affirmative counterplan disposition is that the status quo is an option if the counterplan is dismissed for theoretical reasons but not if the counterplan is simply undesirable as a policy. The affirmative in this case controls the disposition of the counterplan because they can decide to only argue its desirability and thereby stick the negative with it. In the alternative theory, the counterplan always remains subject to the disposition of the negative since they can always abandon it and revert to the status quo.

So then, to answer an often-asked question, what is the difference between a conditional and a dispositional counterplan? My suggestion is that we reserve the term "conditionality" for the traditional view which I have labeled classic, unconstrained conditionality. In a nutshell, the theory of conditionality, as I understand it and as I think it was traditionally understood, is that the negative can advance two or more counterplans in the same round. In contrast, the various theories of counterplan disposition come into play when there is only one counterplan in the round but the status quo is still considered to be an available policy option. There is a longstanding argument to the effect that there is really no distinction between conditional and dispositional counterplans. I think that there is, at least given the way in which I am using the terms, but some degree of confusion is understandable. Dispositionality is perhaps best understood as being a form of conditionality, a logical, limited form of conditionality but a form of conditionality nonetheless. In contrast, we might think of "Conditionality," (with a capital C) as being the theory of multiple counterplans, that is classic, unconstrained conditionality.

That the theory of counterplan disposition is a theory of limited conditionality should be clear. The negative gets one counterplan, just as the affirmative gets one plan. The negative also gets the status quo as a policy option, just as the affirmative gets any legitimate permutation. One may still reject the theory, but it does place clear limits on the number of negative policy options (two). As long as the negative is limited to one counterplan, there can be no slippery slope to more than two options-there is, after all, only one status quo. I have also said that it is a theory of LOGICAL, limited conditionality. In what sense is this the case? The ability to reject both plan and counterplan seems to me inherent in the logic of most decision-making situations. Faced with two competing proposals for change, almost any decision-maker will have the option of rejecting both and sticking with what we have now. To deny the judge the status quo as an available choice defies the logic of rational decision-making as it appears in almost every other real world setting.

Again, perhaps debate should be different. (I will argue in greater depth in favor of logical, limited conditionality below.) The only thing I wish to establish at this point is that there is a principled difference between Dispositionality and Conditionality (one counterplan versus more than one) and that the status quo as a fallback option enjoys a logical status (it is almost always available in the real world) distinct from the theory of classical, unconstrained conditionality. There are many situations in the real world in which we place limits on the number of proposals for change which we consider (prep time, after all, is never really infinite, either within debate or beyond it), but the status quo will almost always remain a logically available option if all of the proposals for change to be considered are deemed inferior. To put it another way, the status quo is always an available benchmark for evaluating any proposal for change, and disallowing it as a choice available to debate judges is a significant departure from the inherent logic of rational decision-making. Perhaps considerations unique to academic debate justify such a move; my point for now is simply that there is an appreciable cost for choosing to do so.

Classic Conditionality Reconsidered:
The Case Against Classic Unconstrained Conditionality

In this section, I want to review the arguments for and against CUC. These are time-honored arguments, but they still appear recurrently in contemporary conditionality/dispositionality debates. Assessing their merits therefore is of both theoretical and strategic interest in its own right, but it also is intended to function as a step toward my ultimate conclusion that logical, limited conditionality (one counterplan plus the status quo) is the best theoretical solution to the conditionality conundrum.

There are four main traditional arguments against CUC, two of which seem relatively principled and two of which are more pragmatic. The first argument is that conditionality violates the principles and practices of good advocacy. This, I think, is the strongest theoretical argument against conditionality, and it is also the most fundamental in that it appeals most clearly to a specific paradigmatic understanding of the debate activity. That said, it should also be noted that "advocacy" is one of those terms which is commonly invoked in debates but which is rarely analyzed. It is often assumed as self-evident that debate is, first and foremost, an exercise in oral advocacy and that sound advocacy practice entails the (almost) complete rejection of conditional arguments. I will suggest below that both of these assumptions are problematic. But for now I just want to attempt to clearly state the advocacy objection to conditional counterplans.

The advocacy position, as I understand it, develops as follows. First, debate is inherently an exercise in advocacy. The affirmative advocates its plan; the negative is obliged to present some kind of counter-advocacy. Clashing advocacies are central to debate.

Second, advocacy training is central to the educational mission of debate.. Lawyers and lobbyists, politicians and political activists all engage in advocacy, and it is a primary purpose of debate to train people to successfully and ethically fulfill these roles in society.

Third, conditional argument is inconsistent with good advocacy. One argument for this conclusion appeals to persuasion research, which it is claimed proves that conditional arguments are unpersuasive. One can see intuitively why this might be the case; arguments taking the form of "My client didn't do it, and besides he was crazy," seem more than a little suspect.

There also seems to be an at least implicit sense that consistent (that is, non-conditional) argument is ethically superior to inconsistent argument. An ethical advocate takes a position and sticks with it throughout.. Shifting from one position to another is a form of sleazy sophistry. This ethical sense that conditionality is sleazy segues into the second major argument against conditionality: that it encourages argumentative irresponsibility. Conditionality allows teams to advance policy positions without (much) cost. If defeated, the position can be simply jettisoned, and the only loss is the time it took to present it. One consequence of this is that teams may feel freer to advance radical, risky, or even repugnant positions, knowing that they won't be stuck with them. Anyone with an even marginally lengthy debate memory can come up with ready examples. Probably my favorite is the "be mean to old people" counterplan (claiming a domestic Malthus net benefit). A more sinister example is a "renew the Holocaust" counterplan (also claiming a domestic Malthus net benefit).

Shifting to the more pragmatic arguments against conditionality, one encounters first and foremost the claim that the time limits within which academic debates occur render conditionality an undesirable option. Less than two hours are allocated for every debate. Even two policy alternatives can't be exhaustively evaluated within that time span, and every additional policy option tossed into the mix will further distract from depth of discussion. "If we had world enough and time" it would be great to consider the full range of possible policies, but we don't. It is therefore best to force each team to do its own "scan" of the available policy options, select one and debate it to the maximum depth possible. A further ramification of the classic, unconstrained conditionality position is that there are really no limits, except for speech times, on the number of conditional counterplans which could be introduced into a given debate round. There is therefore almost unlimited abuse potential associated with this theory.

The fourth and final argument against conditional counterplans is that they legitimate conditional affirmative policy options, including alternative justifications (that is, multiple independent affirmative plans), plan severance, and intrinsicness arguments. The argument for this conclusion is one of simple reciprocity; if the negative can be conditional, the affirmative should also be allowed to be. As a strategic matter, it is sometimes advisable for the affirmative to accept conditionality; affirmative conditionality (including the ability to minor repair away most disadvantages) is almost certainly a more powerful tool than negative conditionality. But as a theoretical issue, the argument proceeds as follows. Negative conditionality legitimates affirmative conditionality. Affirmative conditionality is a terrible idea. (Reasons for this were discussed above.) Thus, negative conditionality should also be rejected.

Answers to the Arguments Against Classic Unconstrained Conditionality

The advocacy argument against conditionality is the most fundamental and philosophical; I therefore want to consider it especially carefully. The first premise of the argument, that debate is intrinsically an advocacy activity, is one which I am willing to accept, albeit with qualifications. But a key question concerns exactly what are the advocacy burdens of the two sides. With regard to the affirmative, I am perfectly content with the traditional answer: it is the affirmative's responsibility to advocate a topical plan with sufficient specificity that productive debate can proceed, and the affirmative should defend the totality of its plan throughout the debate. The affirmative can conceivably add to its plan, through the process of permutation, but it can never subtract from it. (A possible exception would be the dismissal of plan elements which the negative has argued are extra-topical.) But what is the negative's advocacy burden? The conservative answer is that the negative must defend a competitive policy alternative and stick with it throughout the debate. But there is another answer which seems at least broadly consistent with an advocacy-based view of debate: the negative must advocate the rejection of the affirmative plan. In this formula, negative debaters are still serving as (counter)advocates, and they still must clash with the affirmative's advocacy. But conditional arguments can be employed as part of their overall advocacy stance that the affirmative plan should be rejected.

The second premise of the advocacy argument, as I described it, is that it is a central educational mission of debate to teach sound advocacy skills. Again, this is a premise I would accept, with the qualification that this is not debate's ONLY educational mission. What else is there? Some of my colleagues would probably claim that debate is more centrally about teaching argumentation skills in general than with teaching advocacy skills in particular. Others might stress education in such areas of current academic interest as critical theory and postmodern philosophy. My central claim, however, is somewhat different. In my view, a central element of debate's educational purpose should be teaching skills associated with the effective analysis of public policy issues. My point is not that debate should focus on training professional policy analysts. This is a role which some debaters do eventually come to occupy, but it is surely a narrower job niche than law. Rather, I think that debate needs to teach policy analysis skills in the broader sense in which they are relevant to all informed citizens in a functioning democracy. If we wish to be informed citizens, we are all obliged to be effective analysts of public policy, and debate can help a good deal in that process. Is the teaching of policy analysis skills incompatible with the teaching of public advocacy skills? No doubt there is a wide degree of overlap, but there are certain differences in emphasis. Whereas the public advocate may be well advised to take a consistent position and stick to it (flip flops being politically damaging), the informed citizen or the academic social critic is likely to think in a more tentative and hypothetical way about public policy questions.

Faced with a specific policy proposal (like an affirmative plan), s/he might well wish to examine it from a variety of vantage points. S/he might want to engage in a process of mixed scanning-first considering a range of alternatives, then focusing on a smaller number in detail. S/he might want to proceed along the branches of decision tree, considering first, for example, do we want to increase public health services for mental health care?, then considering, if so, what agent could do this most successfully. The citizen and social critic have room to sometimes play "devil's advocate"-for example, by looking to the polar policy extremes and playing them off against the middle. All of these thought processes involve what would be considered "conditional" argument within debate, but they are valuable thought processes, especially for people who don't already have their minds made up, who are approaching public policy questions as thinkers and inquirers instead of as activists and advocates.

My broader point is not that advocacy skills don't matter, nor do I think debate can or should be totally divorced from advocacy. Instead, I am trying to make the more limited point that teaching advocacy (at least advocacy narrowly conceived) is not the only purpose of academic debate. Another important purpose is to teach skills associated with sound policy analysis, skills useful to the citizen and the social critic. And among these skills is the ability to think tentatively and hypothetically, to consider problems from a variety of perspectives, even to entertain mutually exclusive alternatives. Consistency may be a virtue for advocates, but from the standpoint of the thinker and the reflective citizen "a foolish consistency" may be "the hobgoblin of small minds" (Emerson).

But if advocacy and analysis are both important, how does one proceed? The two involve, as I have suggested, differences in emphasis. And it may well be that in the end different judges and debate theorists will simply emphasize one or the other based on their own interests and temperamental dispositions. Still, I do have a suggestion for anyone interested in a middle ground. Hold the affirmative to strict advocacy standards-one plan which they must defend from the beginning to the end. But hold the negative to looser advocacy standards. Still require that they ADVOCATE the rejection of the affirmative plan, but let them do so while incorporating at least limited elements of conditional logic into their arguments. In sum, let them approach the debate as analysts and inquirers, as devil's advocates rather than as the fully committed partisans of a specific policy alternative. This is not an argument for complete critical license, nor is it a defense of negation theory. It is legitimate, I think, to hold the negative to the same degree of policy specificity as the affirmative. I am simply arguing that the negative should get more than one policy alternative.

Another way to think about this is in terms of two different models of advocacy. The traditional advocacy paradigm in debate appealed to the model of the public advocate. The public advocate speaks to a general audience and represents a certain specific interest (or interest group). The public advocate is concerned less with "truth" or with logic than with successfully pleading a case. The public advocate employs relatively simple modes of argument because s/he is addressing the general public. In contrast, we might choose to embrace the model of the academic advocate. Academics also function consistently as advocates. They advocate the acceptance and rejection of certain theories, and sometimes they advocate the acceptance or rejection of certain public policies. But they proceed in a somewhat different way. First, they mainly address a specialized audience. Second, they can focus more on the logical and less on the psychologically persuasive aspects of their arguments. Third, they are also more likely to employ hypothetical modes of reasoning, including mixed scanning, the use of decision trees, reductio ad absurdum logic, and devil's advocacy. In sum, they are more likely to embrace conditional modes of argument than are public advocates.

There are several reasons to think that an academic advocacy model is superior for academic debate. The first is that by definition we are academics; we are not actual public advocates or decision-makers. The position of the academic advocate is therefore more natural, more intuitive, for most of us than is the position of the public advocate. Second, the bulk of our practices in debate harmonize better with an academic advocacy than a public advocacy model. Debate judges are a specialized audience. They can be expected to be more rational in their argument evaluations and less subject to purely psychological influences. Debate argument tends to be fast and technical. It would utterly bewilder a general public audience. Debaters frequently employ arguments which have currency within the academy but nowhere else. Kritiks are an example of this, but this is also true of many policy arguments too radical in their assumptions and implications to appeal to the general public. Since we don't act like public advocates in most of our other debate practices, it is far from clear why conditionality should be the sole exception.

The third premise of the advocacy argument against conditionality is that conditionality is unpersuasive and (potentially) unethical. There is also implicit in this argument a certain appeal to practice, an assumption that real world advocates do not argue conditionally. This, it seems clear, is an oversimplification. In response to a proposal for federal welfare benefits, an opponent of welfare might well say that s/he is opposed to welfare but thinks that if welfare benefits are to be increased it should be done at the state level. This kind of "conditionality" is common even in the world of actual policy making, and it does not seem to be intrinsically unpersuasive. Once again, it seems to be an oversimplification to say that all conditional argument is unpersuasive in all settings and for all audiences. Conditional arguments are, for the most part, logically sound.

Even the argument that "my client didn't do, and besides, he's crazy," presents two logically consistent claims. If your audience is a jury, that may be a bad argument; if your audience is a logician, it could be a good one. Hypothetical argument is a staple of academic, especially philosophical, discourse. If may be used less frequently in the context of public debates, but it is undoubtedly used, sometimes persuasively, even there. In sum, I am suspicious of the logic of completely dismissing an entire mode of argument because there is some research that in some settings and with some audiences it is psychologically counterproductive. Academic debate embraces many practices-including rapid delivery, radical social criticism, and speculatively linked chains of events-which would be unpersuasive for most audiences. In general, our model of argument is far closer to the standards of academic debate than it is to public debate in front of lay audiences. No doubt this has a variety of disadvantages, educational and otherwise. But as long as academic debate assumes in every other way a specialized audience, it seems anomalous to uniquely reject conditional argument because it is unpersuasive to the general public.

The notion that conditional argument is somehow unethical strikes me as even less compelling. Considering several different alternatives does not, on its face, seem morally problematic. Nor does it seem immoral, generally speaking, to modify one's position to some extent over the course of a discussion or a debate. One might even defend such a course as ethically superior to the approach of adopting a dogmatic stance permitting neither compromise nor modification based on new insights.

The argumentative irresponsibility argument strikes me as more persuasive on its face than it is upon consideration. First, it suffers the same problem as all potential abuse arguments. You can reject the repugnant counterplans when you see them without rejecting the vast majority of conditional counterplans which are not repugnant.

Second, it is simply not true that the negative is placed at no risk when they run a conditional counterplan. They make a time investment in the argument. They read evidence and advance analysis which can potentially be used against them. The affirmative may be able to capture the counterplan's net benefit via link turn and argue that it is more of an advantage to the plan than to the counterplan. The counterplan may invoke a permutation which could take out another disadvantage.

Third, while disadvantages can be straight turned, few other types of debate argument entail that same degree of argumentative risk. You cannot turn a harm, inherency, or solvency argument. Attempts to turn topicality (as a reverse voting issue) are unpersuasive. Negatives can almost always get out of kritiks. And even disadvantages are not straight turned all that often. In sum, the argument that the negative should be at HIGH risk for any argument it advances seems untenable.

Fourth, we should not artificially exaggerate argument responsibility. Debaters should be responsible in proportion to the logical weight which an argument carries. If the negative claims that either of two policies is superior to the plan and one of their policies is shown to be inferior, they can still logically win on the other.

Fifth and finally, the fear of repugnant counterplans seems exaggerated. There are natural disincentives to running offensive arguments; they alienate judges and you are unlikely to win on them. In the age of kritiks, extreme violations of political correctness are even more dangerous. And if you want to defend that people dying is good, you can probably do so in the form of a disadvantage without recourse to counterplan as link magnifier.

Of the two pragmatic arguments against counterplan conditionality, the time limits argument is by far the more persuasive. But it too has its problems. It is far more persuasive at the extremes than at the margin. A good debate with twenty alternative negative policy options is difficult to imagine, but there have been plenty of good debates in which the negative has defended two.

Second, it is obvious that speech time within a debate is finite. The time the negative team invests in a counterplan trades off with time they would spend developing other arguments.

Third, it is not clear that the introduction of a counterplan adds a unique level of complexity to the round. The standard affirmative argument is that a counterplan is a complete policy system, raising issues of comparative solvency and desirability as well as a whole gamut of theory issues. The counterplan, it is said, interacts with every other issue in the round, making it a uniquely time demanding negative argument. This argument is probably true-sometimes. But often counterplans are not especially complex. They may make one minor modification in the affirmative plan and generate one narrow net benefit. A small procedural counterplan is likely to be less complex and time consuming than a major meatball disadvantage like Malthus or hegemony. Furthermore, empiricism seems to be against the claim that counterplans are uniquely time consuming. The average number of answers to a counterplan does not exceed the number devoted to a disadvantage, a kritik or a topicality argument. Indeed, it may take less time to answer the counterplan because there are more ways in which the counterplan can be attacked and more of the answers will be unevidenced theoretical objections. As with any argument, you must make a strategic choice concerning how seriously to take a counterplan. If it is a powerful argument, you would presumably want to make many answers. If it is weak or underdeveloped, far fewer answers should suffice.

Fourth, time skews are central elements of debate strategy, at least given the way that the academic debate game is currently played. Teams attempt to skew time with throwaway arguments of all sorts: with low impact or hard to turn disadvantages, with topicality arguments, and increasingly with kritiks. Claims that time skews are voting issues are themselves time skews. I do not find this emphasis on tactical time tradeoffs to be a particularly appealing part of our activity, but given its pervasiveness, it seems a little odd to become uniquely agitated over the time skew potential of conditional counterplans.

Fifth, fewer arguments are not necessarily better. By the logic of maximum depth, the negative could be held to one disadvantage. But if they think they have two or more good disadvantages, we let them be run because there is intellectual value in considering a range of arguments against a given position. The same logic would seem to apply to counterplans. Sixth, with regard to depth of analysis what seems critical is the number rather than the kind of major offensive negative positions. Judges are likely to moan at the mention of "seven off," even if none of them is a counterplan. In contrast, "two off," both of which are counterplans would almost certainly produce a more in-depth debate. In sum, the time limits argument points to a real danger for potential abuse, but it is so selective and situational in its impact that it hardly seems to be a decisive objection to negative conditionality.

The argument that negative conditionality justifies affirmative conditionality strikes me as the weakest of the major arguments against conditional counterplans. At least it is easy to answer if one thinks that the plan is the focus of the debate. If that is the case, then the plan cannot change and affirmative conditionality is disallowed, but the negative can still attack the plan's desirability from more than one policy perspective. This may seem unfair because it is non-reciprocal, but at least a couple of points should be made. The affirmative ability to select a specific case and plan is a huge strategic advantage. The affirmative almost always knows more about the specific policy on which the debate is focusing. And the plan must remain the focus, even if there is more than one counterplan in the round because counterplans are required to compete, that is, provide a reason to reject the plan. Often, the affirmative has a vast advantage in terms of specific knowledge and specific preparation.

Thus, giving the negative more flexibility in terms of its counter-advocacy can easily be considered to be a reciprocal advantage offsetting the affirmative's case selection side advantage. Also, the affirmative isn't really just limited to the plan, the whole plan, and nothing but the plan. Permutations (of which each counterplan is likely to suggest a number) in effect function as additional policy alternatives which can justify an affirmative ballot. Permutations, I will argue below, are inherently conditional arguments, but a legitimate form of logical, limited conditionality. And letting the judge vote for a permutation at the end of the round seems adequate reciprocation for letting the judge vote for a conditional counterplan.

To summarize this section: the advocacy and the time limits arguments against conditional counterplans have considerable merits. (The argument responsibility and the affirmative conditionality arguments are less compelling.) Debate is an activity in which advocacy plays an important part, and in at least some advocacy settings conditional arguments seem to have limited persuasive appeal. The tendency of conditional counterplans to make debate analysis more superficial, no matter how non-unique this may be, is also a valid concern. This is especially a problem if there are no principled limits placed on the number of counterplans which the negative can run. In sum, there are valid, though not utterly decisive objections to classic unconstrained conditionality. I will attempt to argue below that these objections can be adequately answered within theories of counterplan disposition relying on logical, limited conditionality.

The Negative Disposition Option Analyzed

According to the theory of negative counterplan disposition, the negative, like the affirmative is limited to one proposal for change. The status quo, however, always remains a logical option available to the judge. According to this approach, the counterplan is always at the negative's disposition in that they can always concede the counterplan and defend the status quo, even if the affirmative has only attacked its desirability, that is, they have straight turned the counterplan.

The first argument in favor of this position is that it follows from the logic of real world decision-making. A rational decision-maker, confronted with two proposals for change, can almost always reject both and keep things as they are, and indeed, s/he should do so if both changes would be inferior to the policy that now exists.

Second, this form of limited conditionality is also consistent with real world advocacy. Faced with one proposal for change, opponents of a policy frequently offer a counter-proposal. In response to the Republican tax cut proposal, the Democrats will offer an alternative tax cut proposal of their own. But even if the Democrats' plan is rejected, the Republicans' need not be embraced. Democrats can still vote against the Republican plan. Indeed, this posture of proposing an alternative, but still defending the status quo as superior to the policy proposed by one's political opponents is so common that it is essentially business as usual. Thus, rejecting all forms of conditional argument is in fact less in accord with the practice of real world public advocates than is the acceptance of limited conditionality.

Third, the time limits argument against unconstrained conditionality has minimal force against this version of limited conditionality. Each side gets only one proposal for change. The status quo is still a logical policy option justifying a negative ballot; that is, it is a competitive reason to reject the affirmative plan. But there is only one status quo; thus, the number of negative policy options can never exceed two. There is thus no possibility of the extreme abuse potential associated with multiple counterplans.

Fourth, allowing the negative to revert to the status quo is reciprocal with the affirmative's ability to suggest doing both. The "do both" permutation also follows from the logic of rational decision-making. If a rational decision-maker were confronted with two mutually compatible and simultaneously desirable policies, s/he would presumably embrace both. Likewise, if the judge is modeling the behavior of a rational, real world decision-maker, s/he should be able to simultaneously endorse both plan and counterplan, if the two taken together constitute the best policy option. Similarly, s/he should be able to endorse the option to "do neither," since this option equally well follows from the logic of rational, real world decision-making. One could even regard the "do neither" option as a kind of negative permutation of plan and counterplan.

Fifth, normal permutation theory still gives the affirmative more policy options than the negative, even allowing the negative an inherent right to the status quo. There are almost always several legitimate permutations which can be articulated in response to a given counterplan. Thus, in most rounds there will be several policy options (the original plan or any of a number of permutations) justifying an affirmative ballot. Even allowing the status quo fallback, the negative does not reach complete parity. But disallowing even that option constitutes a major failure of reciprocity.

Sixth, the status quo is just one more policy. In a world in which the judge can already vote to do the plan alone, the counterplan alone, or both plan and counterplan, the ability to endorse neither plan nor counterplan, extends the judges options only from three to four.

Seventh, the status quo is a uniquely unproblematic option to add into the mix. The affirmative has already had the whole 1AC to indict the status quo; thus, the status quo has already been subject to reasonably in depth discussion. Disadvantages commonly establish their uniqueness relative to the status quo. The affirmative's initial advocacy is that the plan is superior to the status quo. Not requiring that the affirmative successfully sustain this claim at the end of the round lets them shift their fundamental advocacy stance. Consistent advocacy on the part of the affirmative logically requires them to win both that their plan is superior to the status quo and that their plan (or a permutation) is superior to the counterplan.

Eighth, the negative disposition option constitutes a reasonable middle ground between the extremes of classic, unconstrained conditionality and strict, single option advocacy. Negative disposition helps offset the large side advantage associated with affirmative case and plan selection. It also allows, at least to a limited degree, kinds of hypothetical thinking which contribute to sophisticated policy analysis.

Undergirding this theory of counterplan disposition is a view of debate which I have labeled "policy analysis." Basically, this is the view that the purpose of a policy debate is to find the best policy. At least, this is the best purpose.

Politicians may use debates to gain political victories over their opponents. But in debate we are not active politicians, and we can use debate in a more disinterested way. Why should academic debate be concerned with finding the best policy? Because as citizens we are concerned about finding optimal policies. Because policy makers SHOULD be concerned about finding optimal policies. Because debate should provide training in the processes of rational policy assessment which are essential for a democracy. Training in the logic of finding the best policy option is not the only thing debate should seek to accomplish. It should also train advocates, activists, and academics. But learning how best to pursue the best policy is surely an IMPORTANT thing for policy debate to teach. This does not mean that we can never depart from the logic of rational, real world decision-making. Sometimes other values, competitive or educational, may be more compelling. But it does mean that departures from clear, real world models of rational choice should be undertaken reluctantly and only for compelling reasons.

Academic debate, all too clearly, is not the real world. But it is designed to train us to think, to argue, and to evaluate within the real world. If the logic of sound policy analysis matters in the real world, it should also matter in debate.

Some Further Objections to Negative Counterplan Disposition

In addition to the traditional arguments against conditional counterplans, there is a more recent generation of arguments which has grown up in the age of disposition as opposed to the age of classic conditionality. Most of these are variations on familiar themes, but they are still worth reviewing.

One argument of more recent vintage is that the counterplan is so powerful a tool for the negative that if they choose to run one they should be stuck with it (or at least have the potential to be stuck with it). A counterplan has the potential to completely, or almost completely, negate the affirmative's advantage. It can shift the debate in fair measure away from the affirmative's ground. Given all this, if the negative chooses to run one, its abandonment should not be risk free.

I would not deny, of course, that the counterplan is a powerful tool for the negative, nor that the theory of negative disposition offers strategic advantages to the negative. Every theory offers benefits to one side or the other; strict, single option advocacy is a theory which strongly favors the affirmative. Thus, helping one side or the other is not really a compelling argument against a theory construct. This is especially true if the other side has reciprocal, offsetting advantages (like the power of permutation) or if, in general, side equity remains reasonably in balance. The days of massive affirmative side bias appear to have passed, at least temporarily, but we have hardly entered an era of clear negative preponderance. Thus, arguments of the "pity the poor affirmative" variety seem a bit premature.

Counterplans are powerful tools, but they are not overwhelmingly powerful. Counterplans generally function as defensive arguments; that is, they negate affirmative advantages. But it is rare that a counterplan, operating on its own, provides a sufficient reason to vote negative. Generally, it must be coupled with a disadvantage to serve as a net benefit. Thus, whereas the negative can often win on a disadvantage standing alone (because it outweighs the case), or on topicality, or on a kritik, it is rare that the negative can win just on a counterplan. Since counterplans are not THAT powerful, they need not be subject to special theoretical impediments. A second argument is that the affirmative is too constrained in attacking a counterplan. Conditionality of any sort forces the affirmative to debate in two worlds at once, and this may force them to debate against themselves.

To start with the last point first, it seems clear that the affirmative never HAS to debate against itself. It can always defend the position that its plan (or a permutation) is the best policy option in the round. Affirmative debaters have presumably devoted a whole speech to proving the plan better than the status quo; now all they have to do is to prove the plan superior to the counterplan. If the status quo is still an option, this does place some constraints on exactly how the affirmative attacks the counterplan. They cannot, for example, run a disadvantage to the counterplan which applies even to a lesser degree to the plan if the disadvantage has the effect of proving the status quo superior to the plan as well. But it is not clear to me that the affirmative should be able to make arguments which prove they are inferior to the status quo. This seems to amount to refuting their own advocacy, something that is ruled out in almost any debate rulebook.

As far as debating in two worlds, there is some truth to this argument, but it also seems overstated. The existence of two negative policy options in the same round does create argument interactions which the affirmative must identify and keep straight. But making two policy comparisons at the end of the round instead of just one is not quantum physics. In a world without negative disposition, we have to answer the question, which is best, the plan or the counterplan? In a world with negative disposition, we must answer the question, which is best, plan, counterplan, or status quo? This does not seem monumentally more complex. Besides, debate is supposed to be intellectually challenging. Probably the worst argument against kritiks is that philosophy is too complex to debate about. If we can intellectually cope with postmodern philosophy, global circulation models, and theories of deterrence, I suspect we can cope with the status quo as a fallback option. And the challenges of debating in two worlds at once are reciprocal. The negative also has to keep track of the various argument interactions and make their strategic choices accordingly.

Affirmative Disposition of Counterplans Analyzed

The theory of affirmative disposition is also a form of limited conditionality. Again, the negative is limited to only one counterplan, but within the confines of this theory, the negative cannot concede the counterplan if there are no theoretical objections to it. If the affirmative only proves that the counterplan is disadvantageous and otherwise accepts its legitimacy, the negative is stuck with it. What arguments favor this position?

First, whether or not the counterplan exits or remains in the round is a matter of the affirmative's discretion; the fate of the counterplan is subject to the affirmative's disposition. This makes the counterplan more like a disadvantage. It can never be a risk free argument. But if the affirmative chooses, for its own strategic reasons, to make arguments like "no negative fiat," then they have made the choice to allow the negative to escape from disadvantages to the counterplan.

Second, theory arguments offer rational reasons why the counterplan cannot stay in the round. If there is no negative fiat, or if the counterplan is an abuse of fiat, then it is basically a non-argument, and it has to be dismissed. The same thing is true if the affirmative claims that the counterplan is topical and that counterplans must be non-topical. This means that the counterplan was never fair game for the negative, so it has to be dismissed. If the counterplan is indicted for being partially plan inclusive, the same logic prevails. Of course, one can argue that theory failures should be voting issues. To adequately discuss this question would require another long article. But in a nutshell, I think that this gives too much weight to theory arguments. It prioritizes theory over the substantive issues in the debate. And it rests on an exaggerated notion of the significance of time skews, a matter addressed in terms of the conditionality time skew debate above.

Third, this theory also offers a middle ground. If classic, unconstrained conditionality and strict, single option advocacy are considered the extremes, then both affirmative and negative disposition positions constitute middle grounds. But if negative disposition and strict non-conditionality are considered the extremes (with classic conditionality completely off the map), then the theory of affirmative disposition is likely to seem the most moderate.

The theory of affirmative disposition does possess one unresolved dilemma, which I would like to briefly address. This concerns the status of competition arguments as justifications for the negative conceding the counterplan. Is competition a theoretical burden of the counterplan or is it a question of substantive policy desirability? With net benefits as the overarching standard of competition, it seems that competitiveness is really a matter of substantive desirability rather than a theory question. Through the lens of net benefits, one can say that if a counterplan is undesirable, that is, not net beneficial, then it is not a reason to reject the plan and therefore is noncompetitive. But if being noncompetitive justifies the negative abandoning the counterplan, then the negative can always do so, because all undesirable counterplans are noncompetitive.

What the foregoing logic would suggest is that permutations, because they demonstrate the noncompetitiveness of the counterplan, are not a legitimate basis for the negative conceding it.

There may, however, be another reason why permutations justify negative reversions to the status quo. This is that permutations are themselves a form of conditional argument. Once the affirmative has said that the judge can vote either for the plan or for the plan plus all or part of the counterplan, then they have embraced two policy options. And at that point, the negative is logically entitled to revert to the status quo so that it too has two policy options for which the judge can vote.

I should perhaps note at this point that while I prefer the theory of affirmative disposition to strict, single option advocacy, I vastly prefer the theory of negative disposition to either. The affirmative disposition theory does give the negative a little more room to maneuver than does a theory of strict non-conditionality. And it has for me the not inconsiderable benefit of making counterplan theory issues non-voting issues. But it does not go nearly far enough in terms of following the logic of real world advocacy and analysis or in terms of reciprocating the advantage the prerogative of permutation provides to the affirmative. The arguments in favor of the negative disposition of counterplans apply with full force to the option of strict non-conditionality, but they still apply with considerable force to the option of affirmative disposition.

Are Permutations Conditional?

One of the most powerful arguments in favor of the theory of negative counterplan disposition is that the status quo option is reciprocal with the affirmative's ability to permute. But there is a fairly obvious objection which will doubtless have come to the minds of many readers at this point in the argument. This is that permutations are only a test of competition; they are not a policy option available to the judge at the end of the round. I believe that this view is misguided. I will argue in this section that permutations should indeed be regarded as conditional affirmative policy options and that they are options which the judge can endorse at the end of the round. Furthermore, in light of this, the existence of permutations justifies at least minimal negative conditionality in the form of the retention of the status quo option.

The first point I would like to make is that permutations (or their functional equivalents) are logical and necessary responses to counterplans. Basically, they are our most elegant tool for ensuring that negative counterplans compete. If the affirmative cannot argue for both the plan and all of the counterplan or for the plan and part of the counterplan, then there would be no limit on what counterplans could be run. As noted above, any rational decision-maker who could combine two policies to create an optimal policy package would and should do so.

But if we say the judge can vote to do both, then the judge has two ways to vote affirmative, since it was already assumed that the judge can vote for the plan alone. And this, unquestionably, is affirmative policy conditionality. So if severance, alternative justification, and intrinsicness arguments are all bad, why is this form of affirmative conditionality legitimate? The best answer, I think, is that the affirmative is still defending its full lAC plan (thus there has been no severance of its advocacy), and the permutation (if it is a legitimate permutation) only incorporates policy elements advocated by the negative. (Thus, the permutation adds no new element not found in either plan or counterplan.)

If follows from the logic of rational decision-making that the judge should be able to endorse both the plan and the counterplan. If the judge can endorse the plan by voting affirmative, and if the judge can endorse the counterplan by voting negative, there is no reason why the judge cannot endorse the two policies in combination. This logic is perfectly real world. If Congress has two mutually compatible and mutually desirable bills on its docket, then it can and should pass both. When presented with two policy options the rational decision-maker will almost always have the opportunity to do either option A, option B, or options A and B together. Not to permit the judge this range of choices defies the logic of rational decision-making.

I have just argued that permutations are a form of conditionality grounded in the logic of real world decision-making. They are also a limited form of conditionality. The affirmative may be able to come up with several legitimate permutations to a given counterplan, but the range of permutations is still limited by the content of the counterplan. Of course, if the counterplan is long and complex, there are more potential permutations, but this is a risk the negative assumes when they choose to run a long and complex counterplan.

I have no problem with the position that permutations are a form of logical, limited conditionality and that as such they are options available to the judge at the end of the round. But from the standpoint of strict, single option advocacy, this is not an acceptable solution. Members of the strict advocacy school have therefore gone through a series of theoretical contortions in order to come up with an understanding of permutations in which they are not conditional arguments.

The most popular move along these lines is to claim that permutations are only "tests of competition." This means that the judge does not really vote for them at the end of the round. They merely serve to defeat the counterplan. In a world of strict non-conditionality, this means basically that the negative loses since there is no negative policy option to vote for. But having defeated the counterplan, the permutation can, like a Western hero, now slip quietly into the night allowing the judge to vote for the original plan.

If this seems fishy, it is for good reason. The very language of "testing" arguments should be a dead give away. Within the logic of strict policy advocacy, policy positions must be advocated. A policy position which is not advocated, which is merely a test (like an intrinsicness argument) is a non-argument. The notion of "testing" arguments comes from hypothesis testing, the traditional paradigm which most emphatically embraced conditionality in all of its forms. If permutations "test" counterplan competition, why shouldn't intrinsicness arguments test disadvantages? Why shouldn't conditional counterplans test the plan's necessity and desirability?

Furthermore, the notion that the judge votes for the plan instead of the permutation at the end of the round is basically a conjurer's trick. The judge is de facto voting for the permutation because without the permutation the counterplan would be superior to the plan alone. The decisive comparison, on which the round hinges, is between the counterplan and the permutation, not the plan and the counterplan, or even the plan and the status quo (a comparison which may not even have been addressed since the 1AC).

In my view, the permutation as test notion is a hopeless way out of the problem of permutation conditionality. Tests are conditional arguments. If permutations are tests then permutations are conditional. If the affirmative can employ the (limited) conditionality implicit in permutations, then the negative should be able to employ the (limited) conditionality implicit in the status quo as an always-available logical option.

From a strict advocacy standpoint, there is, however, another way of dealing with this dilemma. This alternative is to limit the affirmative to one permutation and, if they advance a permutation, to require them to abandon their initial plan. Even this does not totally eliminate an element of conditionality from the affirmative's advocacy. It means that the plan does definitely move. And it means that the affirmative is advancing two distinct comparisons, its plan, as superior to the status quo, and the permutation, as superior to the counterplan.

In other words, the affirmative is basically arguing that relative to the status quo, its plan is best, but that if the object of comparison is the counterplan, then the permutation is best. As noted, this still seems to leave an element of conditionality in the affirmative's advocacy. This is seen in the if/then logic of the two comparisons and in the evolution of the plan into the permutation as the policy that the affirmative ultimately advocates. The conclusion that I would draw from this is that once there is a counterplan in the debate, it is practically impossible to avoid elements of conditional argument. But, that said, it seems to me that the approach I have outlined in this paragraph involves the smallest degree of conditionality possible, consistent with attacking counterplan competition. (The affirmative could, of course, be completely non-conditional by never offering a permutation. But in that case they would frequently lose to counterplans which are rather flagrantly non-competitive.)

There may be some strict advocacy theorists who are willing to embrace the just one permutation, either plan or permutation but not both approach. But this theoretically Spartan choice is not in practice very appealing. The reason is that it severely limits affirmative flexibility in responding to counterplans. More specifically, it drastically undercuts the ability to both permute and to run disadvantages to the counterplan. Generally speaking, a disadvantage to a counterplan will also apply to a permutation. (An exception would be a disadvantage to an element of the counterplan not incorporated in the permutation. But this is likely to be a rare occurrence.) So, if the permutation is the affirmative's ultimate advocacy position, the affirmative is running a disadvantage to itself. And even if the disadvantage does not link more to the permutation than to the counterplan (a distinct possibility), unless it links less to the permutation, it can't be a relevant net benefit.

The plan OR permutation option seems to entail the minimum possible degree of conditionality consistent with attacking competition at all. And one could persuasively argue that it is necessary to allow the affirmative at least this much conditionality if it is to have any chance of effectively answering counterplans. But in terms of the logic of rational decision-making, forcing the affirmative to choose between plan and permutation rather than being able to defend a more clearly conditional, either/or approach does not make much sense. If a policy advocate proposes a policy and another policy advocate proposes a counter-policy, it seems perfectly sensible for the first advocate to argue along the following lines. "I think that my opponent's counterproposal is a terrible idea. In terms of its own merits it eminently deserves rejection. But whatever its merits, it is not a reason to reject my proposal. The two a perfectly compatible. So, if you like her idea, you can embrace it, and mine as well." Since this is a perfectly sensible series of arguments in the real world, it is also a series of arguments which we should allow in policy debate. We should not distort the fundamental logic of rational decision-making, which in this case would permit the affirmative to press both horns of a dilemma foreword, in order to limit every vestige of policy conditionality-especially since doing so is a practical impossibility anyway.

Strict Single Option Advocacy Reconsidered

Strict, single option advocacy (complete non-conditionality) has benefits. It does maximize depth of analysis to have two and only two policy alternatives considered in a given debate round. It limits the potential of counterplans to achieve strategic time skews. And it may best comport with real world standards of public advocacy, at least sometimes.

The basic problem with strict non-conditionality is that, in a world of counterplans, it is inconsistent with principles of rational decision-making. Whenever two proposals for change are simultaneously on the floor, two other logical options will almost always emerge-to do both and to do neither. The "do both" option is one which has to be entertained; otherwise there is no real check on non-competitive counterplans. Of course, we can, as a matter of convention, choose to rule out of bounds one or more of these four options. But if the affirmative can defend both its plan and one or more permutations, then reciprocity as well as decision-making logic would seem to dictate that the negative also be able to sustain both the counterplan alone and the "do neither" options. Real world public advocates and rational, real world decision-makers alike employ the kinds of logical, limited conditionality suggested by the status quo fallback and by allowing the affirmative both plan and permutation as policy options. Unless there are compelling reasons, we shouldn't depart from the logic of rational, real world decision-making.

On Fairness and Education

Disputes over debate theory often rest upon disagreements concerning underlying values. At one level, the values appealed to in debate theory argument seem simple enough. Theory claims almost always rest on appeals to some standard of fairness or some standard of education. But both "fairness" and "education" are more complex entities than might appear at first sight. Debate can embrace a number of different educational missions. The debate over conditionality, I have suggested, turns heavily on whether one associates debate's educational mission more strongly with the value of teaching a certain notion of sound advocacy skills or with teaching rational processes of public policy analysis. Other participants in the activity may think it is more important to teach the perspectives of oppositional politics and skills associated with political activism. And of course part of debate's educational mission is still to teach more mundane skills associated with oral communication, research, analysis, refutation, and so on.

A problem in deciding which educational values to stress is that all (or almost all) seem worthwhile. One solution to this is value pluralism, to try to accommodate as many different educational values as possible into the debate activity. I am generally sympathetic to value pluralism, in debate as elsewhere. But pluralism has its limits. For example, there may be certain critical or postmodern approaches and perspectives which do not fit coherently within the framework of public policy advocacy or within the framework of rational policy analysis. And at this point some hard choices may have to be made. If one face of pluralism is the desire to accommodate as many values and perspectives as possible, the flip side of pluralism is the recognition, on occasion, of irreconcilable incompatibilities.

I do not believe there are irreconcilable incompatibilities between the policy advocacy and the policy analysis perspectives. Or, if there are, they exist at the margin. I have suggested what I regard as a middle ground with regard to the conditionality debate. This is to limit each side to the initiation of one policy proposal for change, but to let the affirmative defend any combination (that is, permutation) of those two policies as well, and to let the negative defend both its counterplan and the status quo as superior to the affirmative plan. I have argued that this approach is the most consistent both with standards of rational decision-making and with the actual practice of real world advocates. To restate the argument briefly: when there is only one proposal for change on the floor, there is a relatively easy choice situation-embrace the change or retain the status quo. In a counterplan situation, however, the matrix of choices becomes more complex. When there are two proposals for change on the floor, the rational decision-maker has at least four choices: the initial plan for change, the counter-plan for change, the two together (either in whole or in part), or neither. And since a rational decision-maker would be able to choose from among any of the four options, it significantly distorts the analytical framework of policy debate (viewed as an exercise in rational decision-making) not to allow the judge to vote for any of the four options at the end of the round.

What I strongly suspect, however, is that many readers might follow me this far down the pathway of my analysis and then balk. And the most likely stumbling block is a widespread perception that the positions I have labeled "classic unconstrained conditionality" and "negative disposition of counterplans" are both unfair to the affirmative. To answer this objection, I want to reflect for a bit on the nature of fairness arguments in debate.

Fairness arguments in debate, it seems to me, almost always appeal to some notion of reciprocity. What is fair for one side should be fair for the other. There are also at least two ways in which reciprocity claims can be framed; we might tag these "local" and "global" reciprocity. Local reciprocity focuses on specific theory issues and suggests that both sides should be treated in a reciprocal way in terms of a specific issue. Examples are abundant. Negative conditionality legitimates affirmative conditionality. Affirmative fiat justifies negative fiat. If the plan must be topical, the counterplan should be non-topical. If the affirmative has to defend a specific policy alternative, the negative should have to defend one too. If the affirmative can be counterplan inclusive (via permutation), the negative should be able to be (partially) plan inclusive. Global reciprocity is slightly more complex. A global reciprocity argument is being made when a team argues that one theoretical advantage to one side is properly balanced by another, different theoretical advantage held by the other. In general, local reciprocity arguments seem to be more compelling. It is easier to see how reciprocity is truly preserved if the focus is on a single theoretical issue. Global reciprocity arguments can easily be reduced to the level of "the affirmative wins too much, so you should let the negative do whatever it wants."

Nonetheless, I think that there are valid global reciprocity arguments. For example, a consensus has probably emerged that topicality needs to be a concern only for the affirmative. That is, while affirmative plans must be topical, negative counterplans can be either topical or non-topical. What justifies this apparently non-reciprocal treatment of the affirmative? One justification can be drawn from the logic of rational policy analysis. If the plan is the focus of the debate, then any competitive policy, topical or non-topical, which offers a reason to reject the plan should be grounds for voting negative. In addition, this formulation suggests a global reciprocity argument: the requirement that the affirmative be topical is reciprocal with the requirement that the negative be competitive. Both are basically germaneness requirements. The plan must be germane to the resolution, and the counterplan must be germane to the plan. Though the reciprocity here is not exact, not local, not internal to each specific theory question, it still seems to work.

Though I have disparaged arguments of the "pity the poor negative" or "pity the poor affirmative" sort, there is a certain logic in looking to the bottom line. If one side does win significantly more often, then arguments that a certain theory approach is unfair to that disproportionately advantaged side lose a certain element of plausibility. For a long time, from the 1970s into the 1990s, the affirmative did have a substantial side advantage. Probably affirmative teams won between sixty and seventy percent of the time. When teams won a coin flip, they almost always chose affirmative. In such a world, claims that one should defer to the negative on theory had some appeal. But my sense is that in current debate this side bias has been largely corrected. Negatives now win about as much as affirmatives. Of course, this reflects in part greater theory liberality towards the negative: looser standards of negative conditionality, allowance of plan inclusive counterplans, and so on. It also reflects the emergence of kritiks as powerful negative arguments. Given all of this, the status of global side bias as a basis for theory argument seems somewhat unclear. The fact that the two sides now seem more equal, in terms of competitive results, than they have been for some time tends to cut against the force of this argument. But it is still possible to argue that relatively equity now prevails precisely because of greater theory liberality towards the negative. Neither of these positions strikes me as totally compelling.

What does the conditionality/dispositionality debate look like through the lens of reciprocity? There are several ways in which local reciprocity could be achieved in terms of the conditionality issue. First, we could allow both sides to employ classic, unconstrained conditionality. Since this means everything from alternative justifications and affirmative severance and intrinsicness arguments to multiple conditional counterplans would be allowed, this approach has little current appeal. Second, we could deny all forms of conditionality to both sides. This is the traditional position of strict, single option advocacy. But, as I have suggested, this may be harder than it seems. If counterplans are permitted, then an element of logical policy conditionality seems to inevitably enter debate. And if permutations are allowed (a competitive as well as a logical necessity), then some element of affirmative conditionality seems inevitable. Even employing the strictest possible standard regarding consistency of advocacy, the affirmative must still be permitted to embrace two different advocacies over the course of the debate: the plan in the 1AC and the permutation thereafter.

This suggests to me the third and best way of achieving local reciprocity with regard to conditionality. This is to allow logical, limited conditionality to both sides. Let each side offer only one proposal for change but let each side also defend any logical combination of the two policies. This would mean that the judge can vote affirmative for plan or permutation and negative for counterplan or status quo. This seems locally reciprocal, and this for me largely moots any fairness objections to negative disposition. Indeed, I think that fairness, grounded in the reciprocity of the "do neither" with the "do both" option, is in fact an advantage of the negative disposition approach.

Beyond this, I think that there is also a plausible argument, grounded in global reciprocity, in favor of the negative disposition of counterplans. This is based on the recognition that affirmative case and plan selection options represent a huge side advantage. The affirmative almost inevitably knows more about the particular plan on which the debate is focusing. Since the negative will almost always be at a substantial preparation deficit, they should have more flexibility in attacking the plan from a variety of vantage points. This, of course, is an argument for classic conditionality of counterplans. And unconstrained counterplan conditionality is problematic. A debate in which the only two negative arguments were two conditional counterplans would hardly be superficial relative to most debates which now occur. But unconstrained conditionality places no principled limit on the number of counterplans which can be introduced into the round, and this is a serious problem with the theory. Still, the global reciprocity argument, that affirmative case selection advantages justify some latitude with regard to negative advocacy burdens also seems to have merit. My conclusion from all of this is that global reciprocity considerations at least justify negative counterplan disposition, even if they don't justify unconstrained conditionality.

As I have noted, appeals to reciprocity, either local or global, seem to be the most principled ways of resolving fairness questions. The only other alternative I see is to evaluate matters based on the degree to which a given fairness "whine" resonates subjectively with a given judge. Fairness arguments, I think, commonly suffer from one of several systematic flaws. First, there are arguments that a certain theory or practice is unfair because it benefits one side over the other. This seems less than compelling because adopting the alternative theory or practice would benefit the other side. Second, there are arguments that a certain approach is unfair because it makes debate harder, intellectually more challenging. The problem with this sort of argument is that debate is supposed to be intellectually challenging. If a strategy is hard for the other side to answer, this may mean that it is a good strategy, not that it is an unfair one. Third, there is the recurrent complaint that a certain practice distorts one's time allocation. But time pressures and time allocation choices are intrinsic to debate, at least as it is currently practiced. Both sides systematically seek to distort the other side's time allocation. Even the argument that time skews should be voting issues is commonly advanced in order to secure a favorable time tradeoff.

Appeals to fairness are generally subjective and self-interested. (So, of course, are many other kinds of arguments.) Strategies and approaches which you commonly employ are likely to seem "fair." Those favored by your opponents (especially those you find hard to answer) are likely to be deemed "unfair." I don't foresee the banishment of fairness arguments from debate theory. Nor do I even favor it. Debate is a competitive activity, which means that it is, in part, a game. Unless, as seems unlikely, we agree to write a definitive rulebook, there are going to be disputes about what practices and conventions are fair. I nonetheless think that fairness arguments are overemphasized, both by debaters and by many coaches. It is easy to see why debaters find fairness-based arguments the most compelling. They are in the trenches, and their competitive juices are flowing.

But it is important to remember that debate is not just a game but an intellectual game undertaken for educational purposes. Thus, it seems to me that arguments grounded in the educational values of the activity should generally take precedence over fairness considerations. I am enough of a pluralist to think that we should perhaps be willing to trade off a little education for a lot of fairness. And, of course, we should trade off a little fairness for a lot of education. More commonly, of course, we are not really talking about a whole lot of either. Both fairness and education claims tend to operate at the margin. I would find a debate world without logical, limited conditionality less intellectually satisfying. But debate would almost certainly still be an intellectually challenging, educational activity whatever conventions reign with regard to conditionality. Similarly, someone else might find a debate world in which logical, limited conditionality is acceptable to be less fair than a debate world without it. But a world with limited conditionality is certainly not grossly unfair; with or without limited conditionality, a reasonable degree of competitive equity is likely to prevail.

Given that the effects of a particular theory construct on both fairness and education are likely to be marginal, which claim, fairness or education should be deemed the more compelling? I suggested a moment ago that educational values should generally come first. My rationale for this conclusion is as follows.

First, the educational values of debate persist much longer than the competitive values. What one learns from debate is long lasting; competitive success is relatively ephemeral.

Second, debaters, coaches, and educational institutions invest tremendous amounts of time and money into debate. The investment in the activity seems utterly disproportionate unless it is a highly educational one.

Third, there are lots of good games in the world. The reason we choose to play this one is presumably because we derive unique educational benefits from doing so. Fourth, to borrow a topoi from the kritikers, debate's educational values have an impact in the real world, whereas its competitive values are confined to the round. Some debaters do become actual decision makers, so we hope they will be rational ones. Many debaters become influential, opinion leaders of one sort or another. Almost all debaters participate in some measure in the democratic process, and in doing so it is the knowledge they gained in debate, not their degree of competitive success, that matters.

Maybe the best argument for fairness concerns coming before education is that it is the competitive element which encourages debater participation. If debate were a bad, unfair game, no one would want to play it, and all of its educational values would be lost as well. This is a big impact, but for it to really work as an argument it requires a big fairness link. Effects on fairness, at the margin, are not going to destroy the activity. Nor does it even seem likely that overall participation in the activity is proportional to its fairness. For decades participants accepted a substantial degree of affirmative side bias without engaging in mass exodus from debate. Of course, debaters debate on both sides of the topic, so for half of the time any side bias figures to work in their favor. This certainly limits the ultimate fairness impact of side bias arguments.

The sense of unfairness is undoubtedly greater in elimination rounds, especially those without side equalization, when sides are determined by coin flips. Even this, one might argue, does not make debate unfair-it just makes it more random. Nor is it clear that this randomness discourages participation. As sides become more equal, the skills of the debaters, rather than the luck of the side assignment, become more important. On balance, this is presumably a good thing. But it does have the effect of causing the same teams to win more and more consistently. If randomness lets a wider range of teams do well at tournaments, it might well encourage broader participation in the activity. I am not advocating that we adopt policies to make debate more random. My point is simply that side bias, at the margin, is as likely to encourage participation as it is to discourage it.

Debate's policy-oriented elements have feuded over conditionality for decades. I suppose that this, more than anything, proves that it is indeed a paradigmatic issue, an issue which goes to our core conceptions of the activity. Where core conceptions are at stake, compromise is difficult, and even understanding opposing viewpoints may not be easy. There is value in achieving theory consensus, though consensus doubtless also has its dangers. It would be nice if, thirty years from now, we were no longer engaging in bad tag line conditionality debates (though I suppose that would at least mean that debate, in a recognizable form, had survived). I have suggested that there are options for logical, limited conditionality between the extremes of unconstrained conditionality and unconditional non-conditionality. It would be good if the policy debate community could converge on that conditionality middle ground. There are more important theoretical battles to be fought.

Appendix: On The Conditionality of Kritiks

Are kritiks conditional? My simple answer to this is that some are and some are not. Kritiks often employ the forms of conditional argument. For example, "The affirmative's framework is bad, but even within their framework, their plan is a bad idea." Or, "our opponent's discourse is disgraceful, but even if you like marginalizing, otherizing language, their policy proposal is rotten as well." While arguments like this are conditional in form, they are not necessarily conditional in the strong sense of offering more than one policy option. And it is policy conditionality (conditionality in the strong sense) rather than logical conditionality (conditionality in the weak) sense which is generally controversial.

Kritiks function in a variety of ways. They can function as solvency turns or as indictments of underlying values. They can function as deontological disadvantages or as moral side constraints (as, for example, kritiks of offensive language). They can also function very much like counterplans, attempting to capture much of the affirmative's advantage with a competitive counter-advocacy.

If a kritik is only serving as a harm or a solvency argument, then it is probably not a conditional advocacy. It is when kritiks are functioning most like counterplans that conditional advocacy is most likely to arise. And when the negative, over the course of the round, ends up defending two or more alternatives, it seems reasonable to describe their position as a form of conditional advocacy.

Kritiks become implicated in conditional advocacy in several different ways. First, there might be two or more kritiks, each proposing a different alternative. Second, there might be a critical alternative coupled with a conditional defense of the status quo. Third, there might be a critical alternative coupled with an incompatible counterplan alternative. All of these would seem to be forms of conditional advocacy.

The existence of multiple kritiks is not a sure sign of conditional advocacy. If all of the kritiks are mutually compatible, then they would not seem to be conditional, any more than two or more disadvantages are conditional. The situation becomes a little tricky when the negative says that its alternative is simply "reject the affirmative." In a formal sense, this is just one advocacy. But multiple counterplans could be defended, by the same logic, as non-conditional; they too all ultimately justify rejecting the affirmative. In the case of the multiple counterplans, the global claim "reject the affirmative" contains within it several conditional policy alternatives, each of which is said to independently justify rejection. So, to say that all the counterplans lead to the same final conclusion (reject the affirmative) doesn't really deny that they use conditional policy advocacy as a means of leading to that conclusion.

In terms of kritiks, we probably never really "just reject." We reject something about the affirmative or we reject by embracing an alternative ideology or intellectual framework or worldview. Voting for the kritik means generally voting for its worldview. And that worldview is often conditional, in a strong sense, with other negative arguments. Some examples may help to show this.

Example one. The negative argues that we should reject normative policy argument and that normatively the affirmative plan is a bad idea. This seems conditional. There are two alternatives and the two are mutually exclusive. You cannot normatively reject the plan and at the same time reject all normative legal/policy argument.

Example two. The negative kritiks both statism and capitalism. Even if they say their only alternative is "reject the affirmative," in the case of one kritik you seem really to be voting to reject the affirmative in order to reject statism and in the other case voting to reject the affirmative in order to reject capitalism. If the negative is in effect proposing two alternatives, "reject statism" and "reject capitalism," then it seems they are being conditional. If their alternative is to "reject statism AND reject capitalism," then they are not conditional. Why does this matter? If the negative has to defend the alternatives implied by both kritiks, their advocacy is analogous to a counterplan with two parts. If they can argue to reject capitalism OR reject statism, then it is basically like two counterplans. This is important for the affirmative. Requiring the negative to defend the alternatives implied by both kritiks means that if the affirmative turns one, these turns can be weighed against the other. If the two are operating separately, they probably cannot.

Is kritik conditionality legitimate? One plausible answer is that it is as legitimate as counterplan conditionality. The same, or very similar, arguments speak for and against both counterplan and kritik conditionality. There is, I would suggest, at least one argument to the effect that kritik conditionality is less legitimate than counterplan conditionality. It is sometimes said that conditional counterplans force the affirmative to debate in two different worlds at once. (The negative, of course, has to do so too.) But this seems to be a bit of an exaggeration. Conditional policy arguments at least all operate within the general realm of policy discourse. Conditional kritiks may involve two or more distinct discursive frameworks or worldviews. It is this worldview conditionality which may be deemed uniquely problematic. One reason might be that it is too hard to think, sequentially, within different worldviews.

This does not seem to be a very strong argument. Learning to think within different frameworks is probably a good thing. A better argument, I think, is that if we choose to debate about core underlying assumptions or about fundamental worldviews, we should attempt to do so in as much depth as possible. And especially if these are, as kritikers often argue, the most important things to talk about, then if they are raised, debating them exclusively might well seem warranted. There is also, of course, the problem of the less radical argument contaminating the advocacy of the more radical. You shouldn't eat at the restaurant you are picketing, even if you say the food is bad.

Is there a form of logical, limited kritik conditionality? Probably so. It seems reasonable for the negative to say both "reject statism" and "reject the plan because it makes the evil statist system even worse." This seems closely analogous to the counterplan plus status quo option which I have defended in the main article as legitimate. But once the range of alternatives the negative can choose to embrace goes beyond kritik plus status quo, they would seem to have entered the realm of unconstrained kritik conditionality. And this seems to happen fairly frequently. Negative teams often defend two or more kritiks in the same round, each of which at least implies an independent alternative. Or, a kritik is often coupled with both a counterplan and with a status quo defense. Once again, this would seem to constitute unconstrained, rather than limited conditionality.

Kritiks are not counterplans, but they often function in the same way, as forms of counter-advocacy. Therefore, counterplan theory offers at least a heuristically useful analogue for our thinking about kritiks. Just as counterplan permutations enlighten us as to the mutual compatibility and mutual desirability of policy options, kritik permutations can enlighten us as to the mutual compatibility of the ideas advanced by the two sides in the debate. The plan inclusiveness debate with regard to counterplans should help to illuminate the debate over advocacy-inclusive kritiks. Likewise, what we think about conditionality in the context of counterplans would seem to have relevance to how we think about it in relation to kritiks.

Theory as a Voting Issue: The Crime of Punishment

The early 1980s saw the emergence of what was known, somewhat satirically, as the "punishment paradigm." This was the belief that the judge's ballot should be used to punish the employment of unacceptable theories and practices on the part of debaters. The first article to suggest this approach was written by Doug Sigel, an outstanding debater at Northwestern University, and published in the 1984 DEBATER'S RESEARCH GUIDE. Sigel's article (which I shall refer to as "Sigel 1") was titled, "The Punishment Theory: Illegitimate Styles and Theories as Voting Issues."

In this article, Sigel distinguished between the "old" punishment approach and the "new" punishment practice which he recommended. The old punishment was originated by the judge and aimed at practices which he or she found unappealing. It was interventionist and judgmental and was definitely not grounded in a tabula rasa view of judging. In sum, it was not and is not well calculated to appeal to most debaters. The new punishment, in contrast, was debater initiated. Debaters were to be the ones who made arguments about what practices should be punished, and judges were to judge punishment arguments much as they would any other issue in the debate. Sigel argued that this new approach was needed because debate had been overtaken by a number of unfair and anti-educational practices. (He pointed to unintelligible delivery, multiple conditional counterplans, and counterwarrants as examples.) The reign of punishment was to purify debate of these practices.

This was a plausible theory, but within a year of his first article, Sigel seems to have reconsidered his role as forensic Robespierre. The result of this reconsideration was a second Sigel article (Sigel 2), "Punishment: Does it Fit the Crime?," published in the 1985 DRG. In this second article, Sigel claimed that punishment had become an abuse greater than those which it had set out to correct.

For a time, the position advanced in Sigel 2 appeared to have carried the day. As a matter of theory, punishment arguments became discredited. And in practice their use also declined considerably. Debaters still argued theory, but theory arguments served to justify or reject specific practices within rounds, rather than providing the governing rationale for the decision. Punishment became, like counterwarrants and alternative justification, a largely discredited approach, confined to the fringes of the activity.

This happy condition prevailed until a few years ago, when, like Sauron of Mordor, the Punishment Paradigm began to take shape and to grow again. Of course, I have probably exaggerated a bit. Punishment arguments never totally went away, but they were discredited (in the eyes of a majority of judges) and their use declined substantially. Their reemergence took place by degrees and it was not accompanied by any theoretical manifesto such as Sigel 1. There was simply a tendency for debaters to label more and more theory arguments as "voting issues" and for more and more debates to come down to theory questions. This practice, which we might label "the new new punishment" has been increasingly noted, and it is increasingly condemned, both in conversation and in judge philosophy statements. But insofar as I know, there has not been a systematic effort to reexamine the arguments for and against punishment. It is that task which this article undertakes.

The Case for Punishment

In Sigel 1, there are four major arguments presented in favor of punishment. The first argument was fairness. Certain theories and practices were said to be unfair to opposing debaters. And it is not enough just to reject these practices; they may so skew the round that only voting against the team which employed them can redress competitive equity. The second argument was education. Sigel invoked the view that the judge should serve as an educator. Part of his or her role as an educator is to discourage bad arguments. Unfair theories and tactics may also serve to undercut the educational quality of the debate experience. The third argument was deterrence. Losing debates, Sigel argued, is a powerful inducement for people to change their ways. Debaters are, for the most part, rational animals, and they will respond to strong competitive incentives. Sigel's fourth rationale for punishment was argument responsibility. Punishment with the ballot makes debaters highly responsible for their arguments. And debate, he claimed, should teach debaters to argue responsibly.

The four arguments which Sigel advanced are still, for the most part, the standard reasons advanced in favor of punishment. (At least the first three are. The responsibility argument has proved less popular, and Sigel did not even bother to answer it in Sigel 2.) The new new punishment, however, has given these arguments a somewhat different emphasis.

First, fairness rather than education has become the clear focus of recent punishment arguments. Though pro-punishment debaters regularly invoke both "fairness and education" as reasons to punish, the specific logic they employ is mainly grounded in fairness considerations.

Second, the new new punishment tends to treat theory arguments less as a means of purifying debate of its corrupt practices and more as an ongoing element in the debate game. Theory, indeed, has almost become a new stock issue, an offensive element in many teams' strategic arsenal.

Third, recent punishment arguments seem to place particular emphasis on time skews. This was always a theme in punishment debates, but now the idea that one team has had its time allocation irreparably distorted has become the central justification for punishment.

Fourth, there seems to be increasing emphasis on potential abuse. It is not just what one side may do but what it justifies that is held to be key. Fifth, there is also a claim that punishment arguments are needed to condemn highly abusive practices like evidence falsification and racial slurs. Finally sixth, some have advanced the claim that theory must be a voting issue in order for it to be viable at all within the debate. To win a theory argument at all, it is said, requires so much time investment that other aspects of the debate must be fatally neglected. If debaters (especially affirmative debaters) are to argue theory successfully, it must be an all or nothing issue.

The Constructive Case Against Punishment

The first main argument I would make against punishment is that it exaggerates theory. One view of debate is that it is just a game and that theory is as worthwhile to debate as anything else. In contrast is the view that I would defend: that debate has a substantive intellectual content which it is far more worthwhile to learn about than it is to learn about debate theory. Debate teaches us a great deal about current events and principles of policy analysis, about political theory, political philosophy, and practical politics, about medicine and law, ethics and epistemology. It teaches both problem solving and the criticism of underlying assumptions. And it teaches many other things as well. People disagree about which of these areas of inquiry is most important, but any and all of these subjects are of more intrinsic significance than debate theory.

I write this as someone who finds debate theory interesting. Nor do I think that we can get along without debate theory. Nor should we. Theory is basically a set of meta-arguments, arguments about arguments and about the standards for argument. We could set these standards by authoritative edict (a rulebook) or by convention. But on many theory questions there is widespread disagreement and hence no dominant convention. And in the absence of a prevailing convention there is unlikely to be an authoritative rulebook which could be adopted or accepted. We have come, over the past quarter of a century, to think that these are things which debaters can and should argue about. And I accept this general outlook. But even if we neither want to nor can entirely avoid theory argument, it should not be a central focus of the activity. Yet this is precisely what punishment argument make it. Rather than the criteria for the evaluation of arguments, theories come to be ends in themselves, the pivotal issues on which the debate centrally turns.

This seems misguided. The knowledge gained in debate has many uses in later life, but surely the least useful body of knowledge which debate teaches is debate theory. For those of us who stay in the activity for a long time, it is interesting. We want to sort out in a consistent and satisfying way the principles of our activity. But that still does not make it a very intrinsically important body of knowledge.

I think that we sometimes confuse debate theory with argument theory. I am not arguing that argumentation is not a valid and useful field of thought. And argument theory may intersect with what we commonly think of as "debate theory" at a variety of points. But the vast bulk of debate theory, as argued in competitive debate rounds, really just involves what are appropriate conventions for this particular activity-a contest, sponsored by educational institutions, with a certain format and certain conventions. Are conditional counterplans legitimate? Are plan inclusive counterplans legitimate? Are international counterplans legitimate? Should we assume that the "fiat" of the affirmative plan comes immediately or only after a normal implementation process? Must the affirmative specify an agent? These are the staples of debate theory argument. Especially they are the kinds of issues which most invoke punishment claims. And none of them has particular salience outside the framework and format of contest debate. Of course, it is possible to relate some of these arguments to intellectual controversies beyond competitive debate. For example, a focus on international institutions distinguishes liberalism from realism as foreign policy paradigms. But the debate over international fiat does not draw very heavily on this paradigmatic controversy. And our arguments within competitive debate over the propriety of international fiat does next to nothing to illuminate the liberalism/realism debate within international relations.

Arguments over debate theory are reminiscent of the debates of the medieval scholastic philosophers. Rather than arguing about how many angels can dance on the head of pin, we argue about how many intrinsicness arguments can dance on the head of a conditional counterplan. To Aquinas and company, the relationship of pins and angels was interesting and meaningful. Questions of fiat and conditionality matter to us. But only within the narrower confines of the academic debate activity. Once you leave debate, these issues won't matter to you. So if the focus on punishment serves to make these kinds of arguments more central and other, more exportable forms of knowledge more marginal, then punishment does an intellectual disservice to the students debate is intended to teach.

My second main argument is that, empirically, punishment arguments produce bad, anti-educational debate. Punishment arguments are almost always made badly. They are simply tag lines, especially at the impact level. ("This is a voting issue for reasons of fairness and education.") There are two dominant incentives for labeling an objection to a given theory or practice a voting issue. The first is the "cheap shot" motive. The "independent voter" may get lost in the shuffle, and you may come out with an easy win. I doubt that anyone really thinks that this process of learning to "out tech" your opponent is an important part of debate's educational mission.

Second, by labeling an argument a voting issue, debaters hope of secure a favorable time tradeoff. If an argument is a voting issue, it has to be taken more seriously, even if it is not intrinsically of much substantive importance. Again, in this instance, the punishment argument serves as an element in the tactics of time tradeoff. This is part of the debate game, but it is not a very important part of what debate should teach. As extended, punishment arguments again tend to be a series of tag lines. This is generally true in the negative block, and it is almost always true (because of intense time pressures) in the 1AR and 2NR. If the 2AR chooses to go for a punishment argument, s/he may be more articulate and explanatory. But this generally means that a lot of new arguments are being made, or at least being given flesh from the bare skeleton of assertion, and this raises fairness concerns of its own. Of course, some theory debates are better than others. And I can imagine a world in which theory is debated more clearly and coherently than it generally is in the world of contemporary debate. But the experience of a quarter century of theory debates does not encourage me to think that we will enter that Promised Land any time soon. And "better" theory debates would have to occur in a more thorough and time-consuming fashion than those which occur today. And this would exaggerate the problem of diverting time from more substantive intellectual concerns.

My third argument is that the punishment of voting on theory is almost always disproportionate. To me this seems almost true by definition. Someone advances a "bad" argument. They lose that argument. It is not a decisive argument in terms of the substantive logic of the debate, be that a policy logic, a discursive logic, or a critical logic. But instead of just losing that argument, with whatever logical, limited impact that may have in the round, the team which advanced the "bad" argument suddenly is supposed to lose the whole debate. In other words, every other issue in the round, all of the policy arguments, all of the critical arguments, all of the discursive arguments become moot. They no longer matter and they need not be resolved because one theory argument has been lost.

Beyond my intuitive sense that this is disproportionate, I have two other arguments for why voting on theory is excessive. First, the theories debaters most want to punish are not really that egregious. Punishment claims are most commonly raised against the following practices: conditional counterplans, partially plan inclusive counterplans, permutations against kritiks, extra-topical plan planks, non-specification of agent by the affirmative, and a range of affirmative and negative fiat issues. I personally favor some of these positions, and I oppose others. But I recognize that there are "pretty good" arguments in favor of both sides with regard to all of these issues. In other words, they are all, relatively speaking, "close calls." Or, to put it still another way, there are thoughtful members of the competitive debate activity for whom each of these practices makes sense and others for which they do not. On none of these issues is there a theoretical consensus. And all have been widely employed without "destroying debate." This is not to say that these practices are not fair game for argument. They are. But none is so abusive within the context and conventions of debate as we know it that it needs to be an automatic voting issue. Losing the argument ought to be punishment enough. Nor do we need punishment to deal with theories which the consensus of the activity rejects. The difficulty of winning on counterwarrants or alternative justifications or plan/plan has easily been enough to discourage these practices.

Second, the debate over a given theory issue is, by the end of the round, generally close. Each team has its list of brief, blippy reasons to prefer one theory stance or another. Typically, the two lists are opposed to each other, via a process of grouping, and without clear, on-point clash. Usually both sides drop one or more arguments made by their opponents. So again, typically speaking, judges can almost always find grounds to resolve a given theory debate either way, and they generally do so based on their own biases. In this situation, it once again seems to be enough that one side is penalized by losing the particular theory position. A slight edge to one team over the other shouldn't translate into the critical issue in the round. This is especially true when, as is often the case, the particular theory issue at stake has occupied a fairly small percentage of the total time which the debate occupied.

My fourth major argument against punishment is that it is intolerant. All judges have biases which they are only partially successful at screening out. And perhaps oddly, judges often seem less able to set aside their theory biases than their biases on substantive issues. As I noted above, judges can generally justify voting either way on a given theory argument in most rounds. At least if both sides are putting up a decent fight, this is the case. If a position is conceded, most judges will behave accordingly, though even here there are exceptions. And sometimes there will be such a clear preponderance of argument that judges are unable to find their way back to their own theory predispositions. But with two reasonably skilled teams, it is generally possible to resolve a given theory issue either way, so most judges, most of the time, end up endorsing the theory position which they prefer. This may be an unfortunate fact about judges, and it certainly applies to some judges more than to others, but it is a real tendency. It is hardly controversial to say that judges have biases. But the problem with punishment, in light of this fact, is that voting on theory empowers those biases. Instead of creating a strategic slant, the bias becomes all-decisive. What we should recognize, I think, is that different people can and do legitimately hold different concepts of what debate should be about. If one side appeals to our theory preferences and the other side does not, it is not unreasonable to expect that the side whose views we embrace will win the debate over a particular issue more often. But it is intolerant to rule the other side completely out of order, to decide the whole debate based on this one issue, just because they have gotten on the wrong side of one of our theoretical predispositions.

A fifth problem with punishment is its arbitrariness. Punishment aims at abusive practices. But abuse clearly falls on a continuum. And the line at which sufficient abuse exists to justify a ballot is inevitably arbitrary. Is a ten second time distortion enough to vote on? A ten-minute time distortion? Or where in between? This situation is further complicated by the fact that one never knows just how the team invoking punishment would have allocated its time absent the problematic practice. Claims that they had wonderful arguments which time constraints prevented them from running should be viewed with a good deal of skepticism. Sometimes arbitrary lines must be drawn. But especially when debate's equivalent of the death penalty is involved, that arbitrariness should occasion concern.

My final argument is that punishment snowballs. Once the punishment paradigm is embraced, a likely consequence is what Ross Smith has called "voting issue proliferation." Anything can be labeled as a voting issue. And, indeed, the use of theory as a voting issue has helped to create a class of debate "cheap shot artists" who systematically employ punishment strategies. Losing on cheap shots is infuriating for debaters and coaches, and it is frustrating for many judges to vote on them. They certainly don't make debate a more educational activity. And the teams and debaters who rely on them the most are probably the biggest losers in educational terms.

Answers to Pro-Punishment Arguments

The most frequently invoked of the pro-punishment arguments is that a certain approach irreparably distorted time allocation, destroying the possibility of a fair debate. But the following considerations compel me to believe that time skews are rarely, if ever, voting issues.

First, the attempt to achieve favorable time tradeoffs is a ubiquitous practice in current debate. It is behind the practice of making multiple answers to a given argument. It strongly influences the number of positions the negative team will advance in the 1NC. It is behind the decision to start all of the major negative positions in the 1NC. It dictates how many positions will be extended through the block. It generally controls the decision about whether or not the affirmative should "straight turn" one or more disadvantages. Even the employment of punishment arguments is generally based on the desire to secure a favorable time exchange. It seems silly to single out a few particular instances of this universal practices and say that they are voting issues, when the whole of debate is saturated with strategic time considerations.

Second, forcing teams to make strategic choices does have educational value. Debaters are forced to judge which are their best arguments and be selective about what they will extend.

Third, punishment arguments constitute a self-inflicted coverage injury. It takes time, sometimes considerable time, to argue that a certain approach has distorted your time allocation. If debaters didn't defend punishment, they would have more time to answer other arguments.

Fourth, time skews are often minimal. It is quite common for an issue which occupied literally seconds of the debate to still be tagged as a voting issue. In cases like this, the overall integrity of the round would certainly be maximized by simply rejecting the particular argument rather than the team that made it.

Fifth, teams defending a problematic theory almost always invest some time in advancing that position and in extending it. Time spent answering the time skew argument serves to redress the injury.

Sixth, there are other means of redress rather than the ballot. If some other issue was radically undercovered due to the alleged time skew, the judge could allow new answers on that issue.

Finally, seventh, time skew arguments directed against the affirmative seem especially dubious. The structure of the debate places particular time pressures on the affirmative. The luxury of the negative block should give the negative ample time to answer pretty much whatever the 2AC says. (New 1AR answers do pose a different and more legitimate concern from the standpoint of time allocation issues.)

Pro-punishment debaters frequently appeal to potential for abuse as sufficient grounds for voting against the team which employs a particular practice. This logic is also questionable. First, almost all theory constructs have potential for abuse. Plans can clearly be insufficiently specific, but negative debaters can always call for an ever-greater degree of specificity. Plan inclusive counterplans can focus on such small distinctions that they trivialize the debate. But rejecting all forms of inclusiveness, including either the same agent or the same action would gut the counterplan as a negative strategy. Unconstrained conditionality can be abusive, but strict non-conditionality could be interpreted as disallowing permutations. Most theories become unacceptable if pushed to either extreme, suggesting that it is the practice, not the potential, which is key.

Second, potential for abuse may justify rejecting a particular practice but it does not justify making the use of that practice a voting issue. The point of the potential for abuse position is that there has not really been tangible damage done in this round. In that case, just rejecting the theory rather than voting against the team seems sufficient.

Third, analogies seem to support this conclusion. Drunk driving is illegal because it can potentially lead to death. But we do not treat everyone stopped for drunk driving as a murderer because of that.

Punishment, if is argued, is needed to deal with situations which are unequivocally abusive, such as falsification of evidence, and racial or ethnic slurs. This seems to me to be something of a straw person argument. Perhaps I have just been lucky, but I have not noticed a spate of racial slurs in the debates I have observed. Second, to say that theory should not be a voting issue does not mean that punishment is inappropriate for other practices in debates. Consensus has long supported voting against teams who falsify evidence; this stance certainly does not require you to embrace punishment for theory. Third, the fact that there are truly abusive practices in debate suggests that punishment policies should be reserved for those practices. It may indeed trivialize behaviors which are truly unethical to treat them in the same way as you would an intrinsicness argument or a kritik permutation.

The deterrence argument in favor of punishment does, I think, have some merit. I have no doubt that punishment does, to some extent, deter certain practices. Debaters do respond to competitive incentives. And debaters are more likely to go for theory arguments if they can win the whole debate on them, rather than just securing some strategic advantage. It is hard to seriously go for a theory argument, especially in the 1AR, and still adequately cover the rest of the debate. Conceding all of this, however, there are still a number of factors which mitigate the force of the "punishment is essential for deterrence" argument.

First, consistent rejection of a certain theory construct will deter its use even without the employment of the theory being treated as an independent voting issue. By and large, approaches such as counterwarrants, intrinsicness arguments, and plan/plan are avoided not because of the fear of punishment but because they don't get teams anywhere. With or without punishment, teams are discouraged from running arguments which rarely if ever win.

Second, winning a theory argument generally gives you something important. If you win conditionality is illegitimate, you have defeated the negative's counterplan, presumably an important part of their strategy. (Or you can stick them with it, also an important strategic advantage.) The same if you can win plan-inclusiveness is illegitimate or that the counterplan relies on an unacceptable theory of negative fiat. If you win that the affirmative needs to be held to a specific agent, you will probably be able to sustain your own agent counterplan, also an important strategic gain. If extra-topicality is illegitimate, an important affirmative answer to a disadvantage is likely to be eliminated, even if extra-topicality is not, in and of itself, a voting issue. If you win that kritik permutations are illegitimate, you have defeated what is often the most powerful argument against a kritik. In sum, there are still reasons to argue theory and still advantages to be gained by doing so in a world without punishment.

Third, it is possible to effectively extend a theory argument, even in the lAR, while adequately covering the rest of the debate. This may not be easy, but no one ever said the lAR was supposed to be an easy speech. The difficulty here is that the 1AR is forced to make some tough choices. Your coverage of the issue subject to a theory objection cannot be allowed to distract significantly from other issues in the debate. This means that if you go for theory, say against a counterplan, you will not be able to extend many, if any, of the non-theoretical arguments against the counterplan. If you can undermine a counterplan by winning that it is disadvantageous or that it has a large solvency deficit, you should probably do that, rather than going for theory. Any judge will give weight to the disadvantage or the solvency deficit, whereas not all judges will go for the theory argument. (The debate over the substantive issues is also likely to be more clear-cut than over the theory issues.) But there are times when you don't have any good substantive arguments against a counterplan, and at that point you may need to go for a theory objection. And if you spend the full one to three minutes that you would have spent on the counterplan anyway exclusively on your theory argument, you should have enough time to keep it viable.

Fourth, while I don't doubt that deterrence has some effectiveness, it is probably least effective, in debate at least, against career criminals. Debaters who consistently rely on dubious theories are likely to get very good at defending them. And if they rely heavily on such approaches to win, the chance that they will lose on a punishment argument is less compelling. Their competitive incentive to employ the abusive practice is so strong that it will override the threat of punishment.

Fifth, punishment can deter good arguments as well as bad arguments. Most of us have strategies we support which other segments of the debate community finds offensive. Punishment arguments are made against teams for both specifying and not specifying their agents. They are made against teams both for running and for discouraging plan inclusive counterplans. They are run against kritiks and they are run against politics disadvantages. The pervasiveness of punishment's potential suggests a possible debate Golden Rule: punish not, that ye be not punished.

A final argument sometimes advanced in favor of punishment is that without it certain practices are risk free. There are a number of examples. If extra-topicality is not a voting issue, why not just load up your plan with spikes, then concede them in the face of a rigorous attack? Intrinsicness arguments often take only seconds to make, and they have the potential to completely take out a disadvantage. Absent the threat of punishment, it makes sense to employ them. There is little incentive to specify your agent if the only cost of not doing so is that you will have to defend the most likely agent acting through normal means. Even the claim that topicality is a reverse voting issue has little risk in a world without punishment.

Again, I concede there is a certain amount of truth in this claim, but I don't think that the impact of this argument is great enough to justify the practice of punishment. First, it is possible to exaggerate argument responsibility. Once upon a time, some judges believed that teams needed to justify every plank in their plan. Thus, if one part of the plan was net disadvantageous, the judge would vote negative even if the plan taken as a whole was desirable. (There was also no counterplan to capture the plan's advantageous parts.) Today, this seems silly. Even sillier was an argument once made that teams have a responsibility to extend every argument they make. This suggests to me that we should not artificially weight certain arguments in order to make teams more responsible. Arguments should carry no less and no more than the logical weight they possess in the round.

Second, most of these relatively risk free arguments don't take too much time to answer. An intrinsicness argument can probably be dismissed in about fifteen seconds, and an extratopical provision can be labeled as such just as quickly. (An obviously extra-topical spike doesn't require as much definitional work as does a normal topicality argument.) A brief agent specification argument can be developed in thirty seconds or less. And the argument that topicality is a reverse voting issue is rarely made because consensus agrees that it is such a bad argument. If your opponent chooses to seriously defend its theory construct, you will of course have to further develop your objections. But at that point the theory position your opponent has advanced is no longer cost free. They too have to invest time in defending it.

Dealing with Punishment as a Debater

This section is directed towards debaters who want to avoid playing the punishment game. You may think that this is easier said than done. If your opponents are going to label theory arguments as voting issues, there is a strategic incentive for you to do the same, either reactively or preemptively. I think that this can be a mistake. Once you embrace punishment, then you have largely conceded the ultimate impact to your opponents' punishment arguments. At that point, unless both sides blink and admit there is no in round abuse, punishment logically becomes the framework within which the round should be assessed. The team with the bigger link into its theory abuse argument should win. There are a variety of reasons you may not want the debate to occur within this framework. You may think it places you at a strategic disadvantage because your theory constructs are more subject to indictment than your opponents'. Or they may just be better theory debaters than you are. Or the judge may be predisposed towards their side of the argument. Or you may even think, as I have argued, that punishment arguments are anti-educational and that debate would be a better activity without them. Under any of these circumstances, you will want to keep the debate out of the punishment framework. These are some suggestions for how to do that.

First, be prepared to make anti-punishment arguments. Too often, theory debates are like debates over disadvantages in which all of the arguments focus on the link, leaving a very dubious, overclaimed impact unaddressed. Punishment arguments, at least implicitly, have three elements: you did it, it's bad, and it's a voting issue. Debaters commonly focus on step 2-it's bad-and neglect step number 3, it's a voting issue. It's my contention that arguments against step three, the voting issue rationale, are often the strongest. And there are even instances where you might want to concede step 2 in order to get out of a position. Thus, you might concede that there is no negative fiat or that PICs are illegitimate. This gets you out of the counterplan and back to defending the status quo, a position you may want to be in. But in order to do that you need to take out the impact of the punishment argument-the idea that theory should be a voting issue.

Second, don't empower your opponents' punishment arguments by directly endorsing punishment yourself. As suggested above, this may play right into their hands.

Third, you can and should still make comparative abuse arguments. It is certainly strategic to find things that the other side is doing in the round which you can argue are more abusive than anything you are doing. You can make these kind of theory "link turn" arguments even if you are also advancing the impact takeout argument that theory should not be a voting issue. If, as the debate goes on, you think it is most strategic to go for your theory abuse arguments, you can simply concede the punishment framework (assuming the other team has extended it) and go for the theory turns.

Fourth, if you are worried about theory voting issues against you, try to made some other arguments which will draw anti-punishment claims from the other side. Topicality arguments are often effective in doing this. Affirmative teams will usually make some kind of argument that only actual, in round abuse should be punished. If you concede this, it is likely to help you against opposing punishment arguments, especially on counterplan theory.

Fifth, indict the warrants behind the claim that theory should be a voting issue. (Hint-there usually are no real warrants.) Even if your opponents indicate why your theory approach is unfair or uneducational, they will almost never explain why that is a reason to vote against you as opposed to just rejecting the theory.

Sixth, if you do drop some cheap shot voting issue, appeal for leniency. You might argue that cheap shots trivialize the activity, turning what should be a substantive intellectual inquiry into a game measuring clerical speed and accuracy. Especially if your opponent has a remaining speech in which to respond to your new answers, any unfairness to them is limited. New arguments, in this situation, are arguably less of an affront to debate's core values than the hyper-technical orientation which votes on anything that's dropped and doesn't even give the side confronted with the cheap shot a second chance. Judges will vary in how they respond to this appeal, but more than you think will find it persuasive.

My last suggestion is to try to make theory a reverse voting issue. I advance this suggestion with a certain reluctance. After all, I want fewer debates about theory as a voting issue, and this approach could lead to more. But, if turning punishment tends to discourage its use, the employment of this tactic is consistent with my overall anti-punishment stance.

How then do you make theory a reverse voting issue? Basically, you do this by conceding the punishment impact (step 3 in the punishment argument) and by turning step two. The basic reasoning is the following. Your opponent argues that you are using bad theory and therefore should lose. You will win that in fact you are using good theory. This means that your opponents, having defended the antithesis of your theory, are defending a bad theory. And since theory is a voting issue, THEY should lose.

In practice, this reasoning process can have certain difficulties. Sometimes it is hard to defend a theory practice without at the same time taking out the punishment impact. This is especially true where time skew arguments are concerned. If you get out of the time skew argument by claiming time skews are inherent to the activity, then you can't really argue that the judge should vote against the punishment argument for skewing your time. Also, sometimes you will win that your practice was only minimally abusive; that is, it wasn't a serious enough offense to warrant a ballot. Again, this hardly seems to justify voting against your opponents. Finally, sometimes your turn won't really implicate the impact claimed for why punishment is a voting issue. For example, the other team might claim that you have engaged in an advocacy shift and that advocacy shifts are voting issues. You might win that you haven't shifted your advocacy, but that doesn't prove that the other team has shifted theirs.

Your best chance to make theory a reverse voting issue may well be when there is an education link and impact. If the affirmative argues that you should lose because you ran a plan-inclusive counterplan and plan-inclusive counterplans are anti-educational, you could try to win that plan-inclusive counterplans enhance education. At that point, the affirmative has defended an anti-educational, anti-PIC position, and they should lose.

There are a few other things you can try to do to enhance the viability of your theory reverse voting issue argument. First, you might argue that the other team is the side that claimed that theory is the most important issue in the round. If theory is the most important thing, and we win the theory, then we should win. Second, you might appeal to notions of equalizing risk. If they can win on this argument, they should be vulnerable to losing on it. The analogy might be evidence falsification. If you advance an evidence falsification change and lose it, it is generally considered to be a reverse voting issue. Third, you may be able to use cross-ex to set up this argument. You might ask the other team if they are willing to stake the round on this issue. If they agree, then your turns, if you can win them, do become voting issues. And if they won't agree to stake the round on the issue, then arguably the abuse was not all that great.

In sum I would use this strategy with extreme caution, but it might prove valuable in an appropriate time and place.

Dealing with Punishment as a Judge

This section is intended for judges who have been persuaded to embrace a generally anti-punishment posture. We are still confronted with punishment arguments round after round. How should we respond?

One approach would be simply to never accept theory arguments as independent voting issues. Some rounds will still be decided on theory because theory questions govern the fate of some substantive issue on which the round turns. For example, if the negative strategy depends crucially on a counterplan, and they lose its theoretical legitimacy, they will lose the round, independent of punishment. But judges could simply refuse to ever treat theory arguments as INDEPENDENT voting issues.

The "never vote on punishment" approach has some intuitive appeal in light of many of the reasons I have given for opposing punishment. Still, I think it goes too far. It is such a radical departure from the currently prevailing standards of debate judging, that it may be unfair to debaters. Also, it seems reasonable that debaters have some burden of rejoinder to frameworks like punishment, especially if they are relatively well developed. Finally, this strongly interventionist a judging stance only figures to make you a much-struck judge. Self-marginalization is not the best way to be an influential figure in the activity.

One approach which I do recommend is to hold debaters who advance punishment arguments to relatively high standards of justification. If the round is at stake, if this issue is to be given preference over all others, then the reasons for doing so should be clearly and sufficiently explained. Debate is commonly thought to focus on the plan as an example of the resolution. If focus is to shift from the plan to theory as an end in itself, it seems, once again, that a fairly compelling rationale should be offered.

Punishment arguments are often insufficiently warranted. The standard view of an argument is that it is a claim supported by a reason. Of course, it is difficult to judge exactly where the threshold lies between a minimally sufficient and an insufficient reason. Is "for reasons of fairness and education" enough if those reasons are not articulated? Probably not. But however strict one chooses to be, there should be some standard of sufficient articulation, especially before one makes an argument all decisive.

Second, theory should not be a voting issue if the theory debate itself is either too bad or too close. If all one has on either side is a series of loosely competitive tag lines, then it seems unlikely that the theory debate is clear enough or compelling enough to justify a ballot. Especially is there is some good debating occurring elsewhere in the round, it seems a shame to let the debate be decided based on who, at the margin, did the least bad job of debating the theory. Similarly, if the theory debate is close, the rationale for voting on theory is undercut. If both sides have good reasons for their theory construct, then it seems unlikely that the theory should be deemed highly abusive. And if the theory is only mildly abusive, losing the issue, rather than the whole round seems more appropriate.

Third, I think that judges should allow latitude on dropped, cheap shot voting issues. This is true even if the argument crossed one's threshold for being a prima facie argument. I am not a particular fan of new arguments, though standards in this area have become so loose of late that hardly anyone is too strict. But I am even less a fan of a totally technical view of debate. Proliferating theory voting issues in the hope that one gets dropped is an anti-educational approach to the activity, and judges should do what they can, consistent with our overall commitment to clash, to minimize its effectiveness. In general, I would advocate a one-speech grace period for answering cheap shots. (This assumes that the argument really is dropped due to inadvertence, and that this is not an intentional tactic. I think that one will usually be able to tell the difference.) So, if the team initiating the cheap shot has at least one chance of extend the argument, that should be enough. And even if the answer comes only if the last speech, if you as judge can think of no extension which you would find compelling, the new answer should probably be allowed. That is not to say that this is a perfect solution. There are fairness problems associated with allowing new answers, especially if the drop has led to one strategy being pursued rather than another. The competing equity considerations must be dealt with situationally. But a strict, never a new answer rule tends to make debate too technical, too trivial, and too flowgocentric an activity.

Paradigmatic Underpinnings

Debate paradigms involve core beliefs about the activity. They are fundamental visions of what debate is and should be about. Debate theory once overtly centered on paradigms. They were seen as general models and general philosophies of the activity, and specific theory constructs (such as punishment) were said to derive, deductively, from these broader debate worldviews.

Paradigms receive less explicit discussion today. Rather than overtly stated philosophical principles, they have become for the most part implicit and underlying assumptions. This does not mean that they no longer exist; people still hold core beliefs and frameworks which condition their more specific theory views. But because paradigms tend to be unstated, those underlying assumptions must generally be inferred.

In this article, I have referred at several point to "the punishment paradigm." This is actually something of a joke and misnomer. Originally, the term "punishment paradigm" referred to a type of judge without any clearly articulated philosophy of debate, a judge who simply punished, with the ballot, arguments that he or she disliked. Current defenders of punishment are, for the most part, more sophisticated. They embrace a philosophy of the activity, and their acceptance of punishment derives from that philosophy. Punishment, making theory a voting issue, is not really the core assumption of any paradigm, but it does follow from certain other underlying beliefs.

Inferring unstated premises is a tricky process, especially because different people may accept the same general practice based on different, underlying paradigmatic reasons. This is certainly the case with regard to punishment arguments. There is one set of debate judges and theorists who we might refer to as the "conservative criminologists." They employ punishment in a rule-bound and disciplinary fashion to enforce their own notions of good debate. They tend to view the judge as an educator and to think that the activity has certain rules. It those rules are transgressed, various forms of punishment are properly assessed. This type of judge is something of a dying breed. Debaters in general dislike the disciplinary, interventionist style of judging that this approach entails. And in its prevailing practice, debate judging has become far more tabula rasa, more open to innovation, less judgmental, and more driven by debater practice. This has had both good and bad results.

Most of the judges who currently embrace theory as a voting issue fit into a category that might be labeled "postmodern gamesters." At the risk of caricature, I think that they tend to believe the following things about the activity. First, they tend to think that debate is "just a game." (This is something of a bastardization of the "gaming" paradigm of debate, developed by Snider, which sought to employ the forms and conventions of a competitive game for educational purposes.) Second, there seems to be an assumption among this group that debate is a "pure argument game." This means that the overt content of debate is secondary. The process of arguing, the matching of claim with counter-claim, is what is central. Third, they assume that debate judges should be tabula rasa, that they should try to be blank slates, open to whatever arguments debaters wish to advance.

This "just a game" view of the activity has certain consequences. One is that anything is equally valid to talk about-including the rules of the game. In some ways the rules may even become the central preoccupation because the rules are standards for argument, "pure argument," uncontaminated by real world concerns. Another implication of this view is that fairness concerns become central. If debate is just a game, then making it a fair one is what seems to matter most. The practice of labeling everything a voting issue, which the practice of punishment seems to produce, is also encouraged by this view. Any move in the debate game which has a "voting issue" impact attached is inherently a "strong move." Any move without a direct voting issue impact is a relatively weak move, a move which at best helps to set up the final endgame.

I recognize that this description is a simplification and something of a caricature; I also think that it captures a real tendency within debate. Tabula rasa approaches to debate do have their logic. They tend to be tolerant of different forms of argument, to place the debate in the hands of the debaters, and to encourage direct clash. And certainly the original defenders of tabula rasa judging, like the late Walter Ulrich, believed that it enhanced debate's educational as well as its competitive values.

But tabula rasa, "just a game" approaches tend in the end to be intellectually trivializing. They stress narrow technical skills which have little value beyond the activity. They tend to downgrade the substantive intellectual content of the academic debate. They encourage a focus on esoteric and ultimately trivial aspects of debate theory and to make them "independent voting issues." In sum, they lead to the punishment approach, which it has been the burden of this article to criticize.

What is the alternative? My view is that debate should not be "just a game," but rather an intellectual game played for pedagogical purposes. It is not just a pure argument game, but a game with an overt intellectual content, an explicit subject matter: public policy argument. Of course, we could change the content of debate. But debate , in the form we practice it, has been "policy debate," debate centering on propositions of policy for a long time. And there are good reasons for this. Debate about public policy questions matters; it is an important adjunct to democratic decision-making. Many other fields of inquiry, including ethics, philosophy, and discursive and rhetorical analysis can help to enlighten our views of public policy. But in the end, the logic of policy discourse strongly conditions what arguments are relevant and decisive. Punishment arguments, in the end, move debates away from a focus on policy outcomes. People do things in debate which we dislike. They employ practices which we think are unfair. So there is a natural tendency to want to "punish" them. There is also an understandable competitive incentive to take short cuts to success by labeling peripheral concerns as "independent voting issues." Thus, the appeal of punishment, of theory as a voting issue, is readily explained. But in the end it makes debate a less worthwhile activity. It distracts us from what should be our central concern-learning to think and to argue intelligently about public policy matters.

They are only a test of competition; they are not a policy option available to the judge at the end of the round. I believe that this view is misguided. I will argue in this section that permutations should indeed be regarded as conditional affirmative policy options and that they are options which the judge can endorse at the end of the round. Furthermore, in light of this, the existence of permutations justifies at least minimal negative conditionality in the form of the retention of the status quo option.

The first point I would like to make is that permutations (or their functional equivalents) are logical and necessary responses to counterplans. Basically, they are our most elegant tool for ensuring that negative counterplans compete. If the affirmative cannot argue for both the plan and all of the counterplan or for the plan and part of the counterplan, then there would be no limit on what counterplans could be run. As noted above, any rational decision-maker who could combine two policies to create an optimal policy package would and should do so.

But if we say the judge can vote to do both, then the judge has two ways to vote affirmative, since it was already assumed that the judge can vote for the plan alone. And this, unquestionably, is affirmative policy conditionality. So if severance, alternative justification, and intrinsicness arguments are all bad, why is this form of affirmative conditionality legitimate? The best answer, I think, is that the affirmative is still defending its full lAC plan (thus there has been no severance of its advocacy), and the permutation (if it is a legitimate permutation) only incorporates policy elements advocated by the negative. (Thus, the permutation adds no new element not found in either plan or counterplan.) If follows from the logic of rational decision-making that the judge should be able to endorse both the plan and the counterplan. If the judge can endorse the plan by voting affirmative, and if the judge can endorse the counterplan by voting negative, there is no reason why the judge cannot endorse the two policies in combination. This logic is perfectly real world. If Congress has two mutually compatible and mutually desirable bills on its docket, then it can and should pass both. When presented with two policy options the rational decision-maker will almost always have the opportunity to do either option A, option B, or options A and B together. Not to permit the judge this range of choices defies the logic of rational decision-making.

I have just argued that permutations are a form of conditionality grounded in the logic of real world decision-making. They are also a limited form of conditionality. The affirmative may be able to come up with several legitimate permutations to a given counterplan, but the range of permutations is still limited by the content of the counterplan. Of course, if the counterplan is long and complex, there are more potential permutations, but this is a risk the negative assumes when they choose to run a long and complex counterplan.

I have no problem with the position that permutations are a form of logical, limited conditionality and that as such they are options available to the judge at the end of the round. But from the standpoint of strict, single option advocacy, this is not an acceptable solution. Members of the strict advocacy school have therefore gone through a series of theoretical contortions in order to come up with an understanding of permutations in which they are not conditional arguments.

The most popular move along these lines is to claim that permutations are only "tests of competition." This means that the judge does not really vote for them at the end of the round. They merely serve to defeat the counterplan. In a world of strict non-conditionality, this means basically that the negative loses since there is no negative policy option to vote for. But having defeated the counterplan, the permutation can, like a Western hero, now slip quietly into the night allowing the judge to vote for the original plan.

If this seems fishy, it is for good reason. The very language of "testing" arguments should be a dead give away. Within the logic of strict policy advocacy, policy positions must be advocated. A policy position which is not advocated, which is merely a test (like an intrinsicness argument) is a non-argument. The notion of "testing" arguments comes from hypothesis testing, the traditional paradigm which most emphatically embraced conditionality in all of its forms. If permutations "test" counterplan competition, why shouldn't intrinsicness arguments test disadvantages? Why shouldn't conditional counterplans test the plan's necessity and desirability?

Furthermore, the notion that the judge votes for the plan instead of the permutation at the end of the round is basically a conjurer's trick. The judge is de facto voting for the permutation because without the permutation the counterplan would be superior to the plan alone. The decisive comparison, on which the round hinges, is between the counterplan and the permutation, not the plan and the counterplan, or even the plan and the status quo (a comparison which may not even have been addressed since the 1AC).

In my view, the permutation as test notion is a hopeless way out of the problem of permutation conditionality. Tests are conditional arguments. If permutations are tests then permutations are conditional. If the affirmative can employ the (limited) conditionality implicit in permutations, then the negative should be able to employ the (limited) conditionality implicit in the status quo as an always-available logical option.

From a strict advocacy standpoint, there is, however, another way of dealing with this dilemma. This alternative is to limit the affirmative to one permutation and, if they advance a permutation, to require them to abandon their initial plan. Even this does not totally eliminate an element of conditionality from the affirmative's advocacy. It means that the plan does definitely move. And it means that the affirmative is advancing two distinct comparisons, its plan, as superior to the status quo, and the permutation, as superior to the counterplan. In other words, the affirmative is basically arguing that relative to the status quo, its plan is best, but that if the object of comparison is the counterplan, then the permutation is best. As noted, this still seems to leave an element of conditionality in the affirmative's advocacy. This is seen in the if/then logic of the two comparisons and in the evolution of the plan into the permutation as the policy that the affirmative ultimately advocates. The conclusion that I would draw from this is that once there is a counterplan in the debate, it is practically impossible to avoid elements of conditional argument. But, that said, it seems to me that the approach I have outlined in this paragraph involves the smallest degree of conditionality possible, consistent with attacking counterplan competition. (The affirmative could, of course, be completely non-conditional by never offering a permutation. But in that case they would frequently lose to counterplans which are rather flagrantly non-competitive.)

There may be some strict advocacy theorists who are willing to embrace the just one permutation, either plan or permutation but not both approach. But this theoretically Spartan choice is not in practice very appealing. The reason is that it severely limits affirmative flexibility in responding to counterplans. More specifically, it drastically undercuts the ability to both permute and to run disadvantages to the counterplan. Generally speaking, a disadvantage to a counterplan will also apply to a permutation. (An exception would be a disadvantage to an element of the counterplan not incorporated in the permutation. But this is likely to be a rare occurrence.) So, if the permutation is the affirmative's ultimate advocacy position, the affirmative is running a disadvantage to itself. And even if the disadvantage does not link more to the permutation than to the counterplan (a distinct possibility), unless it links less to the permutation, it can't be a relevant net benefit.

The plan OR permutation option seems to entail the minimum possible degree of conditionality consistent with attacking competition at all. And one could persuasively argue that it is necessary to allow the affirmative at least this much conditionality if it is to have any chance of effectively answering counterplans. But in terms of the logic of rational decision-making, forcing the affirmative to choose between plan and permutation rather than being able to defend a more clearly conditional, either/or approach does not make much sense. If a policy advocate proposes a policy and another policy advocate proposes a counter-policy, it seems perfectly sensible for the first advocate to argue along the following lines. "I think that my opponent's counterproposal is a terrible idea. In terms of its own merits it eminently deserves rejection. But whatever its merits, it is not a reason to reject my proposal. The two a perfectly compatible. So, if you like her idea, you can embrace it, and mine as well." Since this is a perfectly sensible series of arguments in the real world, it is also a series of arguments which we should allow in policy debate. We should not distort the fundamental logic of rational decision-making, which in this case would permit the affirmative to press both horns of a dilemma foreword, in order to limit every vestige of policy conditionality-especially since doing so is a practical impossibility anyway.

Strict Single Option Advocacy Reconsidered

Strict, single option advocacy (complete non-conditionality) has benefits. It does maximize depth of analysis to have two and only two policy alternatives considered in a given debate round. It limits the potential of counterplans to achieve strategic time skews. And it may best comport with real world standards of public advocacy, at least sometimes.

The basic problem with strict non-conditionality is that, in a world of counterplans, it is inconsistent with principles of rational decision-making. Whenever two proposals for change are simultaneously on the floor, two other logical options will almost always emerge-to do both and to do neither. The "do both" option is one which has to be entertained; otherwise there is no real check on non-competitive counterplans. Of course, we can, as a matter of convention, choose to rule out of bounds one or more of these four options. But if the affirmative can defend both its plan and one or more permutations, then reciprocity as well as decision-making logic would seem to dictate that the negative also be able to sustain both the counterplan alone and the "do neither" options. Real world public advocates and rational, real world decision-makers alike employ the kinds of logical, limited conditionality suggested by the status quo fallback and by allowing the affirmative both plan and permutation as policy options. Unless there are compelling reasons, we shouldn't depart from the logic of rational, real world decision-making.

On Fairness and Education

Disputes over debate theory often rest upon disagreements concerning underlying values. At one level, the values appealed to in debate theory argument seem simple enough. Theory claims almost always rest on appeals to some standard of fairness or some standard of education. But both "fairness" and "education" are more complex entities than might appear at first sight. Debate can embrace a number of different educational missions. The debate over conditionality, I have suggested, turns heavily on whether one associates debate's educational mission more strongly with the value of teaching a certain notion of sound advocacy skills or with teaching rational processes of public policy analysis. Other participants in the activity may think it is more important to teach the perspectives of oppositional politics and skills associated with political activism. And of course part of debate's educational mission is still to teach more mundane skills associated with oral communication, research, analysis, refutation, and so on.

A problem in deciding which educational values to stress is that all (or almost all) seem worthwhile. One solution to this is value pluralism, to try to accommodate as many different educational values as possible into the debate activity. I am generally sympathetic to value pluralism, in debate as elsewhere. But pluralism has its limits. For example, there may be certain critical or postmodern approaches and perspectives which do not fit coherently within the framework of public policy advocacy or within the framework of rational policy analysis. And at this point some hard choices may have to be made. If one face of pluralism is the desire to accommodate as many values and perspectives as possible, the flip side of pluralism is the recognition, on occasion, of irreconcilable incompatibilities.

I do not believe there are irreconcilable incompatibilities between the policy advocacy and the policy analysis perspectives. Or, if there are, they exist at the margin. I have suggested what I regard as a middle ground with regard to the conditionality debate. This is to limit each side to the initiation of one policy proposal for change, but to let the affirmative defend any combination (that is, permutation) of those two policies as well, and to let the negative defend both its counterplan and the status quo as superior to the affirmative plan. I have argued that this approach is the most consistent both with standards of rational decision-making and with the actual practice of real world advocates. To restate the argument briefly: when there is only one proposal for change on the floor, there is a relatively easy choice situation-embrace the change or retain the status quo. In a counterplan situation, however, the matrix of choices becomes more complex. When there are two proposals for change on the floor, the rational decision-maker has at least four choices: the initial plan for change, the counter-plan for change, the two together (either in whole or in part), or neither. And since a rational decision-maker would be able to choose from among any of the four options, it significantly distorts the analytical framework of policy debate (viewed as an exercise in rational decision-making) not to allow the judge to vote for any of the four options at the end of the round.

What I strongly suspect, however, is that many readers might follow me this far down the pathway of my analysis and then balk. And the most likely stumbling block is a widespread perception that the positions I have labeled "classic unconstrained conditionality" and "negative disposition of counterplans" are both unfair to the affirmative. To answer this objection, I want to reflect for a bit on the nature of fairness arguments in debate.

Fairness arguments in debate, it seems to me, almost always appeal to some notion of reciprocity. What is fair for one side should be fair for the other. There are also at least two ways in which reciprocity claims can be framed; we might tag these "local" and "global" reciprocity. Local reciprocity focuses on specific theory issues and suggests that both sides should be treated in a reciprocal way in terms of a specific issue. Examples are abundant. Negative conditionality legitimates affirmative conditionality. Affirmative fiat justifies negative fiat. If the plan must be topical, the counterplan should be non-topical. If the affirmative has to defend a specific policy alternative, the negative should have to defend one too. If the affirmative can be counterplan inclusive (via permutation), the negative should be able to be (partially) plan inclusive. Global reciprocity is slightly more complex. A global reciprocity argument is being made when a team argues that one theoretical advantage to one side is properly balanced by another, different theoretical advantage held by the other. In general, local reciprocity arguments seem to be more compelling. It is easier to see how reciprocity is truly preserved if the focus is on a single theoretical issue. Global reciprocity arguments can easily be reduced to the level of "the affirmative wins too much, so you should let the negative do whatever it wants."

Nonetheless, I think that there are valid global reciprocity arguments. For example, a consensus has probably emerged that topicality needs to be a concern only for the affirmative. That is, while affirmative plans must be topical, negative counterplans can be either topical or non-topical. What justifies this apparently non-reciprocal treatment of the affirmative? One justification can be drawn from the logic of rational policy analysis. If the plan is the focus of the debate, then any competitive policy, topical or non-topical, which offers a reason to reject the plan should be grounds for voting negative. In addition, this formulation suggests a global reciprocity argument: the requirement that the affirmative be topical is reciprocal with the requirement that the negative be competitive. Both are basically germaneness requirements. The plan must be germane to the resolution, and the counterplan must be germane to the plan. Though the reciprocity here is not exact, not local, not internal to each specific theory question, it still seems to work.

Though I have disparaged arguments of the "pity the poor negative" or "pity the poor affirmative" sort, there is a certain logic in looking to the bottom line. If one side does win significantly more often, then arguments that a certain theory approach is unfair to that disproportionately advantaged side lose a certain element of plausibility. For a long time, from the 1970s into the 1990s, the affirmative did have a substantial side advantage. Probably affirmative teams won between sixty and seventy percent of the time. When teams won a coin flip, they almost always chose affirmative. In such a world, claims that one should defer to the negative on theory had some appeal. But my sense is that in current debate this side bias has been largely corrected. Negatives now win about as much as affirmatives. Of course, this reflects in part greater theory liberality towards the negative: looser standards of negative conditionality, allowance of plan inclusive counterplans, and so on. It also reflects the emergence of kritiks as powerful negative arguments. Given all of this, the status of global side bias as a basis for theory argument seems somewhat unclear. The fact that the two sides now seem more equal, in terms of competitive results, than they have been for some time tends to cut against the force of this argument. But it is still possible to argue that relatively equity now prevails precisely because of greater theory liberality towards the negative. Neither of these positions strikes me as totally compelling.

What does the conditionality/dispositionality debate look like through the lens of reciprocity? There are several ways in which local reciprocity could be achieved in terms of the conditionality issue. First, we could allow both sides to employ classic, unconstrained conditionality. Since this means everything from alternative justifications and affirmative severance and intrinsicness arguments to multiple conditional counterplans would be allowed, this approach has little current appeal. Second, we could deny all forms of conditionality to both sides. This is the traditional position of strict, single option advocacy. But, as I have suggested, this may be harder than it seems. If counterplans are permitted, then an element of logical policy conditionality seems to inevitably enter debate. And if permutations are allowed (a competitive as well as a logical necessity), then some element of affirmative conditionality seems inevitable. Even employing the strictest possible standard regarding consistency of advocacy, the affirmative must still be permitted to embrace two different advocacies over the course of the debate: the plan in the 1AC and the permutation thereafter.

This suggests to me the third and best way of achieving local reciprocity with regard to conditionality. This is to allow logical, limited conditionality to both sides. Let each side offer only one proposal for change but let each side also defend any logical combination of the two policies. This would mean that the judge can vote affirmative for plan or permutation and negative for counterplan or status quo. This seems locally reciprocal, and this for me largely moots any fairness objections to negative disposition. Indeed, I think that fairness, grounded in the reciprocity of the "do neither" with the "do both" option, is in fact an advantage of the negative disposition approach.

Beyond this, I think that there is also a plausible argument, grounded in global reciprocity, in favor of the negative disposition of counterplans. This is based on the recognition that affirmative case and plan selection options represent a huge side advantage. The affirmative almost inevitably knows more about the particular plan on which the debate is focusing. Since the negative will almost always be at a substantial preparation deficit, they should have more flexibility in attacking the plan from a variety of vantage points. This, of course, is an argument for classic conditionality of counterplans. And unconstrained counterplan conditionality is problematic. A debate in which the only two negative arguments were two conditional counterplans would hardly be superficial relative to most debates which now occur. But unconstrained conditionality places no principled limit on the number of counterplans which can be introduced into the round, and this is a serious problem with the theory. Still, the global reciprocity argument, that affirmative case selection advantages justify some latitude with regard to negative advocacy burdens also seems to have merit. My conclusion from all of this is that global reciprocity considerations at least justify negative counterplan disposition, even if they don't justify unconstrained conditionality.

As I have noted, appeals to reciprocity, either local or global, seem to be the most principled ways of resolving fairness questions. The only other alternative I see is to evaluate matters based on the degree to which a given fairness "whine" resonates subjectively with a given judge. Fairness arguments, I think, commonly suffer from one of several systematic flaws. First, there are arguments that a certain theory or practice is unfair because it benefits one side over the other. This seems less than compelling because adopting the alternative theory or practice would benefit the other side. Second, there are arguments that a certain approach is unfair because it makes debate harder, intellectually more challenging. The problem with this sort of argument is that debate is supposed to be intellectually challenging. If a strategy is hard for the other side to answer, this may mean that it is a good strategy, not that it is an unfair one. Third, there is the recurrent complaint that a certain practice distorts one's time allocation. But time pressures and time allocation choices are intrinsic to debate, at least as it is currently practiced. Both sides systematically seek to distort the other side's time allocation. Even the argument that time skews should be voting issues is commonly advanced in order to secure a favorable time tradeoff.

Appeals to fairness are generally subjective and self-interested. (So, of course, are many other kinds of arguments.) Strategies and approaches which you commonly employ are likely to seem "fair." Those favored by your opponents (especially those you find hard to answer) are likely to be deemed "unfair." I don't foresee the banishment of fairness arguments from debate theory. Nor do I even favor it. Debate is a competitive activity, which means that it is, in part, a game. Unless, as seems unlikely, we agree to write a definitive rulebook, there are going to be disputes about what practices and conventions are fair. I nonetheless think that fairness arguments are overemphasized, both by debaters and by many coaches. It is easy to see why debaters find fairness-based arguments the most compelling. They are in the trenches, and their competitive juices are flowing. But it is important to remember that debate is not just a game but an intellectual game undertaken for educational purposes. Thus, it seems to me that arguments grounded in the educational values of the activity should generally take precedence over fairness considerations. I am enough of a pluralist to think that we should perhaps be willing to trade off a little education for a lot of fairness. And, of course, we should trade off a little fairness for a lot of education. More commonly, of course, we are not really talking about a whole lot of either. Both fairness and education claims tend to operate at the margin. I would find a debate world without logical, limited conditionality less intellectually satisfying. But debate would almost certainly still be an intellectually challenging, educational activity whatever conventions reign with regard to conditionality. Similarly, someone else might find a debate world in which logical, limited conditionality is acceptable to be less fair than a debate world without it. But a world with limited conditionality is certainly not grossly unfair; with or without limited conditionality, a reasonable degree of competitive equity is likely to prevail.

Given that the effects of a particular theory construct on both fairness and education are likely to be marginal, which claim, fairness or education should be deemed the more compelling? I suggested a moment ago that educational values should generally come first. My rationale for this conclusion is as follows. First, the educational values of debate persist much longer than the competitive values. What one learns from debate is long lasting; competitive success is relatively ephemeral. Second, debaters, coaches, and educational institutions invest tremendous amounts of time and money into debate. The investment in the activity seems utterly disproportionate unless it is a highly educational one. Third, there are lots of good games in the world. The reason we choose to play this one is presumably because we derive unique educational benefits from doing so. Fourth, to borrow a topoi from the kritikers, debate's educational values have an impact in the real world, whereas its competitive values are confined to the round. Some debaters do become actual decision makers, so we hope they will be rational ones. Many debaters become influential, opinion leaders of one sort or another. Almost all debaters participate in some measure in the democratic process, and in doing so it is the knowledge they gained in debate, not their degree of competitive success, that matters.

Maybe the best argument for fairness concerns coming before education is that it is the competitive element which encourages debater participation. If debate were a bad, unfair game, no one would want to play it, and all of its educational values would be lost as well. This is a big impact, but for it to really work as an argument it requires a big fairness link. Effects on fairness, at the margin, are not going to destroy the activity. Nor does it even seem likely that overall participation in the activity is proportional to its fairness. For decades participants accepted a substantial degree of affirmative side bias without engaging in mass exodus from debate. Of course, debaters debate on both sides of the topic, so for half of the time any side bias figures to work in their favor. This certainly limits the ultimate fairness impact of side bias arguments. The sense of unfairness is undoubtedly greater in elimination rounds, especially those without side equalization, when sides are determined by coin flips. Even this, one might argue, does not make debate unfair-it just makes it more random. Nor is it clear that this randomness discourages participation. As sides become more equal, the skills of the debaters, rather than the luck of the side assignment, become more important. On balance, this is presumably a good thing. But it does have the effect of causing the same teams to win more and more consistently. If randomness lets a wider range of teams do well at tournaments, it might well encourage broader participation in the activity. I am not advocating that we adopt policies to make debate more random. My point is simply that side bias, at the margin, is as likely to encourage participation as it is to discourage it.

Debate's policy-oriented elements have feuded over conditionality for decades. I suppose that this, more than anything, proves that it is indeed a paradigmatic issue, an issue which goes to our core conceptions of the activity. Where core conceptions are at stake, compromise is difficult, and even understanding opposing viewpoints may not be easy. There is value in achieving theory consensus, though consensus doubtless also has its dangers. It would be nice if, thirty years from now, we were no longer engaging in bad tag line conditionality debates (though I suppose that would at least mean that debate, in a recognizable form, had survived). I have suggested that there are options for logical, limited conditionality between the extremes of unconstrained conditionality and unconditional non-conditionality. It would be good if the policy debate community could converge on that conditionality middle ground. There are more important theoretical battles to be fought.

Appendix: On The Conditionality of Kritiks

Are kritiks conditional? My simple answer to this is that some are and some are not. Kritiks often employ the forms of conditional argument. For example, "The affirmative's framework is bad, but even within their framework, their plan is a bad idea." Or, "our opponent's discourse is disgraceful, but even if you like marginalizing, otherizing language, their policy proposal is rotten as well." While arguments like this are conditional in form, they are not necessarily conditional in the strong sense of offering more than one policy option. And it is policy conditionality (conditionality in the strong sense) rather than logical conditionality (conditionality in the weak) sense which is generally controversial.

Kritiks function in a variety of ways. They can function as solvency turns or as indictments of underlying values. They can function as deontological disadvantages or as moral side constraints (as, for example, kritiks of offensive language). They can also function very much like counterplans, attempting to capture much of the affirmative's advantage with a competitive counter-advocacy.

If a kritik is only serving as a harm or a solvency argument, then it is probably not a conditional advocacy. It is when kritiks are functioning most like counterplans that conditional advocacy is most likely to arise. And when the negative, over the course of the round, ends up defending two or more alternatives, it seems reasonable to describe their position as a form of conditional advocacy.

Kritiks become implicated in conditional advocacy in several different ways. First, there might be two or more kritiks, each proposing a different alternative. Second, there might be a critical alternative coupled with a conditional defense of the status quo. Third, there might be a critical alternative coupled with an incompatible counterplan alternative. All of these would seem to be forms of conditional advocacy.

The existence of multiple kritiks is not a sure sign of conditional advocacy. If all of the kritiks are mutually compatible, then they would not seem to be conditional, any more than two or more disadvantages are conditional. The situation becomes a little tricky when the negative says that its alternative is simply "reject the affirmative." In a formal sense, this is just one advocacy. But multiple counterplans could be defended, by the same logic, as non-conditional; they too all ultimately justify rejecting the affirmative. In the case of the multiple counterplans, the global claim "reject the affirmative" contains within it several conditional policy alternatives, each of which is said to independently justify rejection. So, to say that all the counterplans lead to the same final conclusion (reject the affirmative) doesn't really deny that they use conditional policy advocacy as a means of leading to that conclusion.

In terms of kritiks, we probably never really "just reject." We reject something about the affirmative or we reject by embracing an alternative ideology or intellectual framework or worldview. Voting for the kritik means generally voting for its worldview. And that worldview is often conditional, in a strong sense, with other negative arguments. Some examples may help to show this.

Example one. The negative argues that we should reject normative policy argument and that normatively the affirmative plan is a bad idea. This seems conditional. There are two alternatives and the two are mutually exclusive. You cannot normatively reject the plan and at the same time reject all normative legal/policy argument.

Example two. The negative kritiks both statism and capitalism. Even if they say their only alternative is "reject the affirmative," in the case of one kritik you seem really to be voting to reject the affirmative in order to reject statism and in the other case voting to reject the affirmative in order to reject capitalism. If the negative is in effect proposing two alternatives, "reject statism" and "reject capitalism," then it seems they are being conditional. If their alternative is to "reject statism AND reject capitalism," then they are not conditional. Why does this matter? If the negative has to defend the alternatives implied by both kritiks, their advocacy is analogous to a counterplan with two parts. If they can argue to reject capitalism OR reject statism, then it is basically like two counterplans. This is important for the affirmative. Requiring the negative to defend the alternatives implied by both kritiks means that if the affirmative turns one, these turns can be weighed against the other. If the two are operating separately, they probably cannot.

Is kritik conditionality legitimate? One plausible answer is that it is as legitimate as counterplan conditionality. The same, or very similar, arguments speak for and against both counterplan and kritik conditionality. There is, I would suggest, at least one argument to the effect that kritik conditionality is less legitimate than counterplan conditionality. It is sometimes said that conditional counterplans force the affirmative to debate in two different worlds at once. (The negative, of course, has to do so too.) But this seems to be a bit of an exaggeration. Conditional policy arguments at least all operate within the general realm of policy discourse. Conditional kritiks may involve two or more distinct discursive frameworks or worldviews. It is this worldview conditionality which may be deemed uniquely problematic. One reason might be that it is too hard to think, sequentially, within different worldviews. This does not seem to be a very strong argument. Learning to think within different frameworks is probably a good thing. A better argument, I think, is that if we choose to debate about core underlying assumptions or about fundamental worldviews, we should attempt to do so in as much depth as possible. And especially if these are, as kritikers often argue, the most important things to talk about, then if they are raised, debating them exclusively might well seem warranted. There is also, of course, the problem of the less radical argument contaminating the advocacy of the more radical. You shouldn't eat at the restaurant you are picketing, even if you say the food is bad.

Is there a form of logical, limited kritik conditionality? Probably so. It seems reasonable for the negative to say both "reject statism" and "reject the plan because it makes the evil statist system even worse." This seems closely analogous to the counterplan plus status quo option which I have defended in the main article as legitimate. But once the range of alternatives the negative can choose to embrace goes beyond kritik plus status quo, they would seem to have entered the realm of unconstrained kritik conditionality. And this seems to happen fairly frequently. Negative teams often defend two or more kritiks in the same round, each of which at least implies an independent alternative. Or, a kritik is often coupled with both a counterplan and with a status quo defense. Once again, this would seem to constitute unconstrained, rather than limited conditionality.

Kritiks are not counterplans, but they often function in the same way, as forms of counter-advocacy. Therefore, counterplan theory offers at least a heuristically useful analogue for our thinking about kritiks. Just as counterplan permutations enlighten us as to the mutual compatibility and mutual desirability of policy options, kritik permutations can enlighten us as to the mutual compatibility of the ideas advanced by the two sides in the debate. The plan inclusiveness debate with regard to counterplans should help to illuminate the debate over advocacy-inclusive kritiks. Likewise, what we think about conditionality in the context of counterplans would seem to have relevance to how we think about it in relation to kritiks.