Guidelines for Reaching a Reasoned Judgment

Mark Battersby and Sharon Bailin

1. Introduction

When one begins to survey the work on conductive argument, two surprising facts emerge. One is that so little has been written on conductive arguments; the second is that much of what has been written has focused on establishing their existence. One would have thought that even a cursory observation of how arguments are conducted in all areas of life would bring to an observer’s attention not only the existence of such arguments but their ubiquity. Making judgments based on both pro and con considerations is a common phenomenon in numerous domains, as Govier notes: “In my experience they [conductive arguments] naturally occur in law, philosophy, interpretive studies—and in fact in any area, including science, in which there are reasons for and against, or ‘pros and cons’ which we must consider in order to make a judgment on an issue” (Govier 1999b, p.160).

That the ubiquity and importance of conductive reasoning has not been sufficiently recognized may be a function of its “messiness.” Conductive arguments do not fit traditional argument patterns. The premises neither entail the conclusion nor do they support the conclusion in an unambiguous way as some of the “premises” (or anti-premises, as some have called them) actually adduce reasons that count against the conclusion. Indeed, Johnson makes the point that conductive arguments are not easily identified as arguments by either Formal Deductive Logic or positivism (Johnson 2000a, p.92).

Given their lack of conformity to traditional argument patterns, the appraisal of conductive arguments has become a central issue. As they are non-conclusive arguments, one cannot specify the criteria for their formal validity, as Wellman points out. And since they involve reasons against as well as for the conclusion, the problem arises as to how to weigh the various considerations and counter considerations, especially as such a weighing will be dependent on subject matter (Wellman 1971, pp.61–62; Govier 1985, p.261). For these reasons, some theorists have concluded that, “It is difficult to give any general guidelines about appraising conductive arguments” (Govier, p.260). Wellman argues, in fact that, although it is meaningful to refer to the validity of conductive arguments, the only way to establish such validity or lack thereof is by thinking the argument through and feeling its logical force (Wellman 1971, p.79).

It is our belief that there are some general guidelines which can be offered with respect to doing conductive reasoning well and that these guidelines give rise to a set of criteria for identifying inadequate conductive arguments. It is such guidelines and criteria that we elucidate in the remainder of the paper.

2. What Do We Mean by Conductive Reasoning?

Before proceeding with that task, it is necessary to clarify how we are using various terms and to delineate the focus and scope of our project. Our focus is not on the structure or assessment of particular conductive arguments per se but rather on the enterprise of conductive reasoning. By conductive reasoning we are referring to the process of comparative evaluation of a variety of contending positions and arguments with the goal of reaching a reasoned judgment on an issue. We adopt this terminology and focus for a number of reasons.

The first is clarity. What are generally referred to as conductive arguments are most likely themselves constructed of competing arguments which may offer reasons in support of a particular claim, objections to and critiques of arguments offered, or responses to objections. We will call the collection of all arguments in a piece of conductive reasoning a case, and individual arguments, simply arguments. A case, then, is made up of a collection of arguments whose conclusion is intended to support a particular judgment on the issue in question. Let us illustrate with an example (taken from our textbook on inquiry). This dialogue takes place following an extensive evaluation by the two protagonists of the various arguments commonly offered for and against capital punishment.

Phil: You know, Sophia, we’ve looked at a lot of arguments and information on capital punishment, but I think that the conclusion is becoming obvious to me. The weight of arguments clearly points against capital punishment.

Sophia: What made you come to that conclusion?

Phil: Well, it’s pretty clear that there’s little evidence to support the deterrence argument.

Sophia: Agreed.

Phil: And the incapacitation argument is really “overkill” (sorry about that) since the same result can be achieved by less drastic means.

Sophia: Agreed again.

Phil: The issue of cost actually supports the con side since it turns out that capital punishment is much more expensive than life imprisonment.

Sophia: Right again.

Phil: I think that there is something legitimate to the retribution argument in terms of the desire for justice. But retribution can be achieved with life imprisonment. You’ve also convinced me that with capital punishment, we risk an even greater injustice, that of possibly executing an innocent person.

Sophia: I’m with you.

Phil: Then there’s the fact that capital punishment is discriminatory.

Sophia: True.

Phil: And then we’re left with all the moral problems of the state killing some of its citizens and, in particular, some of its citizens who are innocent. That’s a very strong argument against capital punishment.

Sophia: Especially since there are alternatives.

Phil: And given the worldwide trend toward abolition, supported by important organizations like the United Nations, the arguments for capital punishment would have to be very strong to counter that.

Sophia: Which they’re not.

Phil: So, all in all, I have to agree with the abolitionists—we should not have the death penalty (Bailin and Battersby 2016, pp.235-236).

This dialogue may be seen as exemplifying a conductive argument in the usual sense, offering as it does a number of independent reasons in support of a conclusion as well as addressing objections and counter considerations. As noted above, however, this presentation of the case is preceded by considerable reasoning in the form of an evaluation of individual arguments and a comparative weighing of considerations that leads to the making of this conductive “argument.” Cases are often presented in this way—as summaries of conductive reasoning, using primary claims to support a judgment without an explicit statement of the arguments that provide support for these claims. But good conductive reasoning involves a deeper process of inquiry in which the credibility of primary claims is based on an assessment of the arguments that provide support for these claims and in which competing considerations are explicitly weighed and balanced. This is the process in which Sophia and Phil have been engaged previous to this dialogue. It is this entire process of comparative evaluation and weighing which is the focus of our interest, and not simply the resulting “argument.”

Conductive arguments, in the usual sense, can vary considerably in subject matter and complexity. Both the preceding argument regarding capital punishment and the argument: “I’m tired, but I should go to the store anyway because we need bread” have the structure of a conductive argument. Our focus, however, is on the former. We are interested in the pro and con reasoning which takes place in complex and controversial situations, the kind of comparative evaluation we make in actual contexts of disagreement and debate.

Another reason for focusing on conductive reasoning is our commitment to the view of argumentation as dialectical. According to Blair and Johnson, “To say that argumentation is dialectical … is to identify it as a human practice, an exchange between two or more individuals in which the process of interaction shapes the product” (Blair and Johnson 1987, p.46). Our primary focus is on what will make this process a successful one, thereby leading to an adequate product, i.e., a credible reasoned judgment.

3. Features of Conductive Reasoning

The guidelines and criteria we offer arise from the particular features of conductive reasoning.

The first characteristic of import here is that the appropriate goal of conductive reasoning is not the making of a conclusive argument but rather the making of a reasoned judgment. By a reasoned judgment we mean not simply a judgment for which one has reasons, but a judgment for which one has good reasons, reasons which meet relevant standards. A piece of conductive reasoning can, at best, offer good, but not decisive, reasons to support a conclusion over its competitors. Thus arriving at a reasoned judgment will require an examination and weighing of the reasons offered on different sides of an issue and the balancing of various considerations.

No survey of arguments will be exhaustive, however. The possibility always exists that additional reasons and arguments will be put forward which might affect the outcome of the reasoning. Thus the judgment that is the outcome of the conductive reasoning process is always provisional and open to further examination. In addition, because this type of reasoning takes place in complex contexts with dimensions of uncertainty, there may be more than one judgment that is defensible given the context. For these reasons, conductive reasoning needs to be seen in the context of an ongoing process of critical inquiry.

Conductive reasoning takes place in many domains (as mentioned above). It is common in practical reasoning (Hitchcock 2000, pp.5–8) and in social theory and history (Govier 1985, p.260), but can also take place in virtually any domain, including art interpretation and criticism and scientific inquiry. In addition, reasoning about many contested issues will involve a range of types of considerations (for example, factual, ethical, practical). As a consequence, a variety of different types of considerations will often need to be taken into account in conductive reasoning and the criteria of specific domains of inquiry will often play an important role.

An important feature of conductive reasoning (of the kind which is of interest to us) is that it takes place in the context of a dialectic, of a historical and ongoing process of debate and critique, of competing views and the give-and-take among them. Reasons and arguments have been offered on various sides of the issue in question, objections have been raised to many of the arguments, responses have been offered to some of the objections, and alternative views have been put forth. This constellation of reasons, arguments, objections and responses constitutes what Johnson calls the dialectical environment (Johnson 2007). Having knowledge of the dialectical environment surrounding an issue is central to the enterprise of arriving at a reasoned judgment (Bailin and Battersby 2009). In addition, knowledge of the history of the debate can be of assistance in determining which arguments are salient and should be considered, which are considered strong, and which are considered defeated and why.

In addition to this dialectical context, we have identified several additional aspects of context that we believe are relevant to conductive reasoning by playing a role in the determination of both the significance and the weight of reasons. One is the state of practice, which refers to the current situation with respect to the issue at hand (e.g., is there currently capital punishment in the jurisdiction under discussion, and if not, when was it defeated and why). Knowing where the force of current practice and opinion lie can help us to understand what alternative views are up against and whether (and to what extent) any of these views bears the burden of proof. Knowledge about the intellectual, social, political, and historical contexts that surround an issue can contribute to our understanding of the assumptions that lie behind various positions and why people might hold them. Hitchcock’s observation that students’ problems with conductive reasoning are due in part to a “lack of background knowledge to generate a full enough range and detail of competing considerations” (Hitchcock 2000, p.7) points precisely to the centrality of this kind of contextual knowledge.

The dialectical nature of conductive reasoning implies that the process will be dynamic. Particular arguments are often modified or reframed in response to criticism and objections, and these modifications may in turn result in a reframing of the objections, and so on. As Zenker points out, for example, “Typically, some premises appear only in response to and sometimes integrate an opponent’s objections successfully” (Zenker 2007, p.2). In this spirit, Wohlrapp argues against a view of (non-deductive) argumentation in terms of a sequence of isolated inference steps and for a view in which “premises and conclusions of an argumentation form a ‘retroflexive’ system of mutual support” (Wohlrapp 1998, p.342). One implication of this dynamism is that weighing arguments cannot be simply a matter of placing competing arguments on a metaphorical balance scale because arguments will often change in the process of reasoning. Conductive reasoning will need to give attention to the modification, reframing, and synthesizing of arguments.

Because conductive reasoning involves the comparative weighing of reasons on various sides of an issue and because there will often be good reasons supporting different judgments, how strong the prevailing case is in comparison to the other cases will vary. Thus the strength of the judgments warranted by particular instances of conductive reasoning will vary as well. This feature of conductive reasoning points to the need to apportion the confidence of one’s judgment to the strength of the reasons.

4. Guidelines for Conductive Reasoning

In what follows we offer guidelines for conducting conductive reasoning, and then use these guidelines to identify various fallacies in conductive reasoning that one might see either in the process of reasoning or in a case instantiating such reasoning. These guidelines arise from the dialectical and contextual nature of conductive reasoning reviewed above.

4.1. Appropriately review the “dialectical space,” i.e., identify the relevant arguments and the history of the debate

As noted above, in coming to a reasoned judgment, the first task is to conduct an appropriate inquiry into the relevant arguments, including a review of the history of the debate. In addition to providing information regarding the salience and strength of various arguments, the history of the debate provides a context without which it may be extremely difficult to understand some arguments. For example, the problematic nature of the debate in British Columbia and then across Canada about the wisdom of a carbon tax was largely the result of the fact that most citizens were unaware of the dialectical context of the debate. For many, it was just another “tax grab” by the government with the puzzling and suspicious feature that the money was being returned to the taxpayer. Most simply did not understand the economic argument about carbon being an externality (a cost that was not fed through the market) that needed to be woven into the price structure of goods if there was to be an economically rational revision of the use of carbon fuels. The context was not simply global warming, but an extensive debate that had occurred among policy theorist about how best to implement incentives for reduction of carbon use.

4.2. Consider the full variety of objections to the various arguments and responses to the objections

The arguments pro and con about an issue which are the substance of conductive reasoning need to be identified and evaluated along with their associated objections. It is worth noting that there are at least two kinds of objections to individual arguments that provide the support for the primary claims. We suggest using the following terminology. An under cutter is a critique of an argument offered in support of a primary claim. This critique could attack the premises of the argument or the inference to its conclusion. The goal of an under cutter is to show that the conclusion of the argument is poorly supported so that the argument’s conclusion cannot serve as a credible primary claim in support of the case’s judgment. For example, an under cutter for the argument that capital punishment deters would be evidence showing that jurisdictions which eliminated capital punishment did not experience an increase in murder. Another kind of objection to an argument in a case is a specific counter—a countervailing argument or claim meant to provide a countervailing consideration to a particular primary claim. The claim that capital punishment will inevitably result in the execution of people who are innocent is directly countered by the argument that all socially useful practices have downsides which must be accepted; on this view the execution of innocents is just something that society needs to accept in order to have appropriate punishments for first degree murders. These two kinds of objections directed at particular primary claims differ from general counter arguments or con arguments. Con arguments provide a different kind of objection. For example, the argument that capital punishment is a barbaric practice inappropriate to civilized countries is not an argument directed at any particular argument for capital punishment. Rather, it is a general countervailing consideration or con argument.

4.3. Evaluate individual arguments according to relevant criteria

Since the very concept of conductive reasoning involves marshalling both pro and con arguments and relevant objections, one of the primary requirements for reaching a reasoned judgment is that relevant pro and con arguments must be evaluated (just as one would do with any argument). This is not an assessment of the “weight” to be given to a certain claim in the case, but rather an assessment of credibility of the primary claim given the review of the supportive arguments and objections. For example, one could evaluate the arguments for the claim that capital punishment does not deter using the usual criteria for causal claims in the social sciences. Alternatively, one could point out that the appeal to a police chief’s opinion is a fallacious appeal to authority. One could evaluate the evidence for the claim that historically, innocent individuals have been executed and for the claim that it is unlikely that this problem could be eliminated (the latter by appealing to historical evidence, legal scholars, etc.). Finally, one could evaluate the moral argument that capital punishment is the only appropriate punishment for certain kinds of murder—this would require a largely philosophical inquiry.

4.4. Establish the burden of proof and standard of proof required

One role that the consideration of context plays is to help identify, where appropriate, which side bears the burden of proof and the relevant standard of proof required. In scientific inquiry, the burden of proof bears on any novel theory or on claims counter to well established views. Science is inherently conservative in this way. In the political situation, those who argue for change in statutes or other political arrangements inevitably bear the burden of proof. But the standard here can clearly and reasonably evolve. After fifty years of widespread usage of marijuana and at least some scientific studies, the claim that it is relatively harmless (not harmless, but compared to alcohol…) is widely accepted and therefore claims of relative harmlessness would not bear the same burden of proof as they might have in 1960. Even more decisively, the argument that prohibition would not stop usage seems so obvious that it could almost be assumed in the argument. Returning to the capital punishment debate, the claim that capital punishment is not an effective deterrent against murder is now the accepted view of criminologists and anyone arguing for a deterrence effect would bear the burden of proof.

4.5. Assess possibilities in light of alternatives

Part of the assessment of particular arguments should involve consideration of whether there are better alternatives to the position being advocated. For example, with respect to the claims that capital punishment is necessary for both incapacitation and retribution, the existence of the less morally troubling alternative of life imprisonment provides an alternative that weakens the force of those claims. In addition, since the goal of conductive reasoning is reasoned judgment, an inquirer should not be restricted to only considering alternatives that have been put forward in the past. Part of the resolution of a longstanding controversy may well be to consider totally different alternatives rather than trying to decide which of given alternatives is worthy of support. On the question of the legalization of marijuana, for example, there is a wide range of alternatives to consider. While California contemplates legalization of marijuana, many other jurisdictions are considering just decriminalization for possession, or as in The Netherlands, its sale in only certain “coffee bistros.”

4.6. Take into account the relevant range of considerations

Because reasoning about many contested issues will involve a range of types of considerations (for example, factual, ethical, practical), it is important to ensure that one has taken into account the appropriate range of considerations when attempting to make a reasoned judgment. So, for example, in examining the issue of whether we should eat meat that comes from factory farms, it would be important to take into account both factual considerations about the conditions of animals kept on these farms and ethical considerations regarding whether humans have a moral obligation to animals. In inquiring into the debate over the raising of the minimum wage, it would be important to consider not only statistical information, but also the differing assumptions about equity and merit which are inherent in different positions in the debate. In dealing with public policy issues, it would be important to consider ethical as well as instrumental considerations, ends as well as means, costs as well as benefits, and long term as well as short term consequences.

4.7. Take into account and consider a variety of perspectives

The goal of reasoned judgment involves the attempt to make a decision or assessment from an ideal observer’s or “objective” point of view, striving for the “view from nowhere” as the regulative ideal. Striving for this ideal involves attempting to look at an issue from many relevant perspectives—e.g., in a moral dilemma trying to see the perspective of both the moral actor and those of the victims or beneficiaries of the action. One might consider, for example, the controversy surrounding Peter Singer’s advocacy of the euthanasia of disabled babies. Many disabled groups argued that he had failed to consider their perspective (McBryde Johnson 2003).

4.8. Consider differences in how issues, arguments, and reasons are framed

Opposing arguments are frequently characterized by different ways of framing or setting up the issue. Particular ways of framing may slant an inquiry in a particular direction and reframing may affect the outcome of the reasoning. Kahneman and Tversky (1982) have shown, for example, that the question of whether a decision is framed in terms of losing lives versus saving lives has significant impact on the way most people make the decision. As another example, a deontological approach to moral issues would frame a moral dispute quite differently than would a consequentialist perspective. The debate over carbon tax provides yet another illustration of the significance of framing. After the public outrage in British Columbia about the carbon tax, a PR person suggested that what the government should have done was to reframe the issue from a proposal for a tax increase to a proposal for “tax shifting,” i.e., shifting taxes from income tax to carbon producing activities. The carbon tax would not be a tax increase but a tax shift, which would be more acceptable and intelligible to the average citizen, a claim supported by poll results (Barrett 2008). Recognizing differences in framing can often help one to understand the assumptions underlying opposing arguments and thus to be in a better position to comparatively evaluate them. It also opens up the possibility for a mediation of frames that may lead to a judgment that incorporates the strong points of the opposing views.

4.9. Recognize and attempt to incorporate/synthesize strong points from different positions

Good reasons often do not reside entirely in one or other of the conflicting views. Thus it is important, in arriving at a reasoned judgment, to recognize the valid points in each view. The best-justified judgment is often one that incorporates the strong points in opposing views. In the dialogue, for example, our participants acknowledge that the need for deterrence, incapacitation and retribution are legitimate concerns, but they argue that they can all be addressed through life imprisonment.

4.10. Appropriately weigh and balance different considerations, values, and arguments

A central aspect of arriving at a reasoned judgment involves weighing the various reasons pro and con. Although there will likely be some differences in views about comparative weight, it is possible to justify one’s assignment of weight and to criticize reasoning for inappropriate weighting (see below for a detailed discussion of weighing).

4.11. Consider whether one’s own personal convictions and experiences may be coloring one’s judgment

Since we are focused on the process of arriving at a reasoned judgment, there is a requirement for the participant(s) in this process to be aware of their own biases and prejudices. Increasingly convincing research has demonstrated the difficulty people have in making reliably rational judgments. Efforts, including the sharing of discussion with others, identifying one’s perspectives and biases, and avoiding the more common generic biases such as representativeness (thinking individual events or experiences are representative of what generally happens) and confirmation bias (seeking only instances that provide support for one’s view) can all serve to make it more likely that one comes to a judgment which is truly reasoned. One key strategy to avoid bias in one’s judgment is to give due attention to evidence and arguments that counter one’s own point of view. As noted above, we have built such considerations into the process of inquiry, so there is already an important check on confirmation bias, although other biases may need to be addressed with different strategies. An awareness of the historical basis of one’s views and those of others can also help to undermine an inappropriate confidence in one’s views.

4.12. Make a judgment at the appropriate level of confidence—apportion one’s judgment to the strength of the reasons

Part of rational self-awareness involves assessing how much confidence one should have in one’s judgments given the arguments that one has reviewed. It may be that one can conclude with considerable confidence that capital punishment should not be used by a state, but as current debate about what to do about global warming or the debate about the causes of obesity show, not all judgments can be made with the same degree of confidence, even though there may be an urgent need to act on such judgments. Judgments of the likelihood of descriptive factual claims present one sort of problem, but any judgment about what to do must also take into account future states of affairs that are usually less certain than judgments about current states of affairs. And finally, while there are some accepted general moral principles, their application in particular cases, especially ones where accepted principles conflict, inevitably creates significant uncertainty. The unpredictability of the future means that almost all significant actions need to be based on judgments that are at best less than fully confident. In our text we suggest the following table as a guide.

Judgment and Confidence

A very confident judgment is warranted when the weight of reasons clearly supports the judgment.

A reasonably confident judgment is warranted when the weight of reasons strongly supports the judgment but there are still strong countervailing considerations.

A tentative judgment is warranted when the weight of reasons is not overwhelming but is supportive of one position, and we can make a judgment on balance.

A suspended judgment is warranted when the reasons for different positions are closely balanced or when there is insufficient evidence to make a judgment (Bailin and Battersby 2016, p.243).

5. Failures of Judgment

Our focus to this point has been on offering guidelines for reaching reasoned judgments. We also believe that these guidelines can furnish the basis for identifying certain kinds of problems in particular pieces of conductive reasoning, or cases. A given case can be evaluated in terms of the extent to which it deals with, or fails to deal with, the relevant considerations for reaching a reasoned judgment. We have termed the failures “failures of judgment.” As is the case with traditional informal fallacies, failures of judgment are most useful in identifying bad arguments rather than in specifying good ones. We propose that proffered cases are inadequate to the degree to which they fail to take into account the various relevant considerations. The following is a description of the failures of judgment which we have identified.

• Failure to undertake a comprehensive examination of the various competing arguments

Since reaching a reasoned judgment involves a comparative evaluation of the various reasons and arguments on an issue, the failure to take into account any of the significant arguments on the issue constitutes a serious defect in a case.

• Failure to give appropriate consideration to the burden of proof

Failing to determine where the burden of proof lies or misplacing the burden of proof may result in an inappropriate determination of how much evidence is needed to make a case or of when a case has been made successfully.

• Failure to consider the uncertainty of claims

Taking claims as certain where the evidence in support of the claim is not, in fact, compelling may result in making an unjustified judgment or making a judgment with a greater degree of confidence than is warranted.

• Failure to consider alternative solutions or possibilities

The strength of a case can only be evaluated in light of the alternatives available. Ignoring possible and plausible alternatives would be a ground for criticism of a given case.

• Failure to consider objections

Because argumentation is dialectical, any reasoned case, in addition to offering arguments, must also respond to any known and important objections. Failure to do so significantly weakens the case.

• Failure to consider implications

Many cases concern decisions about what to do. However correct an action may appear on the basis of the arguments provided, failure to consider consequences (typically unintended consequences) significantly weakens the case.

• Failure to consider a range of considerations

Judgments which fail to take into account relevant considerations are faulty for that reason.

• Biased framing

Too narrow framing of an issue or argument, or framing in a way that slants the discussion toward a particular perspective may exclude the consideration of other possibilities and thus bias the judgment.

• “Either-Or” framing

Given that many issues have more than two sides, and that there are often intermediate possibilities between two opposing positions, viewing all issues in terms of ‘either-or’ – as a choice between two opposing positions, can oversimplify issues and result in a failure to recognize other, possibly more reasonable possibilities.

• Inappropriate weighting

This problem consists in giving undue weight to certain aspects of an issue when making a judgment.

• Making a judgment at an inappropriate level of confidence

Asserting a judgment with more or less confidence than warranted by the strength of the reasons constitutes another fallacy of judgment.

6. Weighing and Balancing Considerations

A central notion in discussions of the evaluation of conductive reasoning, including our own, is that of weighing. Whatever guidelines may be offered, in the final analysis, reasons pro and con must be weighed in order to reach a reasoned judgment. Yet weighing is a metaphor that is difficult to cash out in the context of arguments, as numerous theorists have pointed out. Is it possible to quantify the weight or strength of various reasons or arguments? And if it is not, then does the notion become so vague as to be of little use or so subjective as to be devoid of evaluative purchase (Koch 2007c).

It is our view that weighing (which we take as the process) is a meaningful, if imperfect, metaphor, and that although weightings (which we take as the products of weighing) are not quantifiable and will sometimes be the object of disagreement, they are nonetheless not (or not primarily) subjective. Weightings can be justified (or criticized) by appeal to objective factors and considerations, for example by appeal to certain widely shared values and principles. Moreover, arguments can be evaluated in terms of both the likelihood that they are true and the support or weight that they give to the judgment. An argument which, if its conclusion is credible, gives considerable weight to a judgment will add little or no weight if it is doubtful. In the court context, for example, an argument that shows that the accused had a good alibi will largely exclude a conviction, whereas if the alibi is in question, the weight it provides is greatly diminished. On the other hand, the fact that an individual has a credible motive adds relatively little weight given that many people may have motives for committing a certain crime.

The excerpt of the dialogue on capital punishment quoted earlier can be used to illustrate some of these aspects of weighing. It is important to bear in mind, however, that a considerable amount of discussion regarding the relative weight of various arguments has already taken place before this dialogue occurs (e.g., Phil: But you’ve convinced me that with capital punishment, we risk an even greater injustice…) and that this discussion process has been a dynamic one, with some of the arguments being modified or reframed in the course of the reasoning that has led to the presentation of the case that we see in the dialogue excerpt.

When reviewing their previous evaluation of individual arguments, Phil and Sophia agree that two of the pro capital punishment arguments, the deterrence and cost arguments, do not hold up—their conclusions are not justified. They are refuted by under cutter arguments and thus are given no weight. However, in addition to the likelihood that the conclusion is true or credible, the arguments can also be assessed with respect to the amount of support (positive or negative) they provide for the case for capital punishment. And each of the deterrence and cost arguments, if they had been credible, would have added different amounts of weight to the case for capital punishment. If capital punishment really did serve as a significant deterrent to murder, that would be a strong argument in its favor, grounded as it is in the widely shared value of saving the lives of innocent people. Even if it were true, however, that the costs are greater to incarcerate for life than to execute, that would not constitute a strong argument in light of the moral objections to capital punishment because of the prima facie presumption that moral issues should generally trump instrumental issues such as cost.

Another of the pro capital punishment arguments, the incapacitation argument, is recognized as sound in the sense that it is true that dead murderers cannot murder again. Nonetheless it is seen as a rather drastic way of removing murderers from circulation given there are other possibilities and so is not a very strong argument for capital punishment. Thus this argument is weakened by a specific counter argument that there is a less morally troubling alternative, life imprisonment, that can achieve the same goal. The retribution argument, on the other hand, is seen as based on strong grounds—an appeal to justice, which is a widely shared value and one that is inherent to any legal system. Nonetheless, although the legitimacy of the appeal to justice is recognized, the weight of the argument as a justification for capital punishment is lessened because life imprisonment can be seen as an alternative which also meets the demand for justice. For both the incapacitation and retribution arguments, then, their weight in the debate is reduced because of the existence of less problematic alternatives.

The likelihood of executing innocent people is viewed by our two inquirers as a very strong argument against capital punishment, indeed as a consideration which overrides most other considerations, appealing as it does to a very strongly held value (not to kill innocent people) and a basic principle of the law (not to punish the innocent). It is true that any system of punishment will have errors no matter how good a job the system does in trying to avoid them. It is, however, crucial to the strength of the argument that some executions (and other long-term incarcerations) have been shown to have been erroneous. So the execution of innocent people is not just a theoretical possibility or an exceedingly rare occurrence. The frequency of such occurrences and the racial bias evident in many cases, in at least some locations, add to the strength of the argument. The weight given to this argument must still be seen as comparative, however, in that, if it could be shown that capital punishment were a significant deterrent and that it would thereby prevent many more innocents from being murdered than would be victims of system error, a much stronger case could be made for the practice. Because of the comparative nature of these evaluations, numbers, if credible and appropriate, may be significant.

We can also see how an appeal to the question of burden of proof is used to help determine how strong the arguments on various sides would need to be in order to prevail. In this case, the worldwide trend toward abolition sets up a burden of proof on the retentionist side. The determination of burden of proof is less pivotal in this case as the anti- capital punishment arguments have been judged to be considerably stronger, but it can be decisive with respect to issues where the reasons on each side are judged to be more evenly balanced. Consider the criminal trial situation where the burden of proof is clearly on the prosecution. The failure of the defense to decisively undermine the prosecutor’s argument should not result in the defendant being convicted since all the defense needs to do is show that there is a reasonable doubt about the guilt of the accused.

One aspect of weighing that is illustrated in the proceeding is that an important ground for justifying weightings is an appeal to widely shared values and principles. The extent of agreement in this regard should not be underestimated. There would, for example, be widespread agreement that the legal system should instantiate principles of justice; that moral considerations should generally take priority over cost considerations; that the state executing innocent people is extremely ethically problematic. Some of these values and principles are built into various domains and related to the “point of the practice.” It is, for example, a basic principle in law not to convict or punish an innocent person. The alleviation of suffering is a foundational value in medicine. Education is grounded in the learning of the child. Weightings can be legitimately justified in terms of such values and principles, and judgments can be rationally criticized which exhibit inappropriate weightings. We would, for example, be justified in criticizing an educational policy if it was seen to value administrative efficiency over the learning of the child.

An excellent example of this aspect of weighing is provided by Allen in his paper discussing Canada’s Rape Shield decision (Allen 1993) where he cites an excerpt from the opinion of one of the judges regarding the exclusion of possibly prejudicial evidence in rape cases:

When, however, prejudicial evidence is for the defence, the prejudicial effect it would have if admitted must substantially outweigh its probative value before a judge can exclude it. This is because a free and democratic society attaches great importance to the principle that an innocent person must not be convicted (p.106).

Here we have both an explicit statement of a central principle that ought to be appealed to in legal decisions and a judgment about the appropriate weighting of considerations in a particular case based on this principle.

There can, of course, be disagreement, even at times deep disagreement, about the relevant or primary considerations, as seems to be the case, for example, in the abortion debate. It is often the case, however, that there will be agreement on the considerations but disagreement over how to prioritize them or how they play out in particular instances. In the rape shield decision cited by Allen, for example, a dissenting opinion by another of the judges argued that the excluded sexual history evidence “is either irrelevant or so prejudicial that its minimal probative value is overwhelmed by its distorting effect” (Allen 1993, p.107). In this case there is agreement regarding the principles that are relevant, i.e., prejudicial effect vs. probative value, but disagreement about their relative weighting in this particular context. As another example, amongst people toward the left of the political spectrum, there are those who support a carbon tax because they believe that it would have a positive impact on the environment while there are others who oppose it because they believe that it would have a negative impact on economically disadvantaged individuals. Although both groups value both the environment and economic equality, they prioritize these values differently with respect to this particular issue. These differences in judgment may be based to some extent on differences in how the likelihood or the severity of the various possible outcomes is assessed or how the short term versus the long term costs and benefits of the different policies are calculated. But these are differences for which one can offer justifications and about which one can reason.

Another example of an explicit discussion of weighing can be seen in a groundbreaking paper by Cornfield (1959) in the context of the early debate about whether smoking caused lung cancer. Cornfield argued that despite the fact that researchers could not provide a good biological model (i.e., animal experiments) to demonstrate the link between smoking and lung cancer, in this case that criterion should not be given the weight it usually receives in epidemiological reasoning. His argument was that, since smoking exhibited very strong correlations and a strong “dose relationship” with lung cancer, these facts and the fact of the lack of credible alternative explanations for the data should be taken as adequate to establish a causal link between smoking and lung cancer. This was one of the first successful arguments in epidemiology since the late 19^th century to subordinate the biological account to the results of large-scale statistical results.

We take these examples to show that there is a role for a rational examination of weightings and the considerations that lie behind them. In this regard, the two opinions cited in the rape shield case (or the argument by Cornfield) can be seen as models for the role of the justification of weightings. Such an explicit justification of weightings puts them forward for scrutiny in the arena of public reason where they can be the basis for deliberation by others and for ongoing inquiry. Since weighing is a dynamic process, there is always the possibility that arguments and even issues may be reframed, resulting in the dissolution of a disagreement over how values or considerations have been weighed. An example would be a public policy debate, initially framed in terms of the competing rights of various parties being reframed in terms of the welfare of the community. Such a process will not necessarily lead to agreement among the interlocutors, however. But unless and until the issue is considered settled, any evaluation made can be seen as a moment in and contribution to on ongoing public process of reasoning about the issue by others as well as ourselves.

References

Allen, D. 1993. “Relevance, Conduction and Canada’s Rape-shield Decision.” Informal Logic 15, 2: 105-122.

Bailin, S. and M. Battersby. 2009. “Inquiry: A Dialectical Approach to Teaching Critical Thinking.” In Argument cultures: Proceedings of OSSA 8 CD-ROM, edited by J. Ritola. Windsor, ON: OSSA.

______. 2016. Reason in the Balance: An Inquiry Approach to Critical Thinking, 2^nd Edition. Cambridge, Mass: Hackett; 2010. 1^st Edition, McGraw-Hill Ryerson.

Battersby, M. 2006. “Applied Epistemology and Argumentation in Epidemiology.” Informal Logic 26, 1: 42-61.

Cornfield, J. et al. 1959. “Smoking and Lung Cancer: Recent Evidence and a Discussion of Some Questions.” Journal of the National Cancer Institute 22: 173-203.

Govier. T. 1985. A Practical Study of Argument. Belmond, CA: Wadsworth.

______. 1987. Problems in Argument Analysis and Evaluation. Providence, RI: Foris Publications.

______. 1999. The Philosophy of Argument. Newport News: Vale Press.

Hitchcock, D. 2000. “Statement on Practical Reasoning.” Retrieved from http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.29.6897&rep=rep1&type=pdf , downloaded Jan. 16, 2010.

Johnson, R. H. 2000. Manifest Rationality: A Pragmatic Theory of Argument. Mahwah, NJ: Erlbaum.

______. 2007. “Anticipating Objections as a Way of Coping with Dissensus.” In Dissensus and the Search for Common Ground, CD-ROM, 1-16, edited by H.V. Hansen, et al. Windsor, ON: OSSA.

Kahneman, D., P. Slovic and A. Tversky. 1982. Judgment Under Uncertainty: Heuristics and Biases. Cambridge: Cambridge University Press.

Kock, C. 2007a. “Is Practical Reasoning Presumptive?” Informal Logic 27, 1: 91-108.

______. 2007b. “Norms of Legitimate Dissensus.” Informal Logic 27, 2: 179-196.

McBryde, Johnson, H. 2003, Feb. 16. “Unspeakable Conversations.” New York Times. Retrieved from http://www.nytimes.com/2003/02/16/magazine/unspeakable-conversations.html?ref=petersinger

Wellman, C. 1971. Challenge and Response: Justification in Ethics. Carbondale: Southern Illinois University Press.

Wohlrapp, H. 1998. “A New Light on Non-deductive Argumentation Schemes.” Argumentation 12: 341-350.

Zenker, F. 2007. “Complexity Without Insight: Ceteris Paribus Clauses in Assessing Conductive Argumentation.” Retrieved from www.frankzenker.de/downloads/ZENKER_ALTA_2007.pdf, downloaded Jan. 15, 2010.