![[Table of Contents]](/icons/up.gif){kind=icon}
According to NRCSL Associate Director James Voss, argument, too, is an inferential process by which the principles, facts, and conclusions of a given discipline are supported. The grounds for judging whether arguments are sound are the discipline's standards for evidence. In social science and other "ill structured" disciplines, says Voss, assertions and conclusions are often based on beliefs and supported by informal, or everyday, reasoning. Voss studies the construction, evaluation, and justification of informal arguments in such fields in an effort to develop an empirical understanding of informal reasoning and to determine the best ways for cultivating it in the classroom.
In social and political science, arguments are as ubiquitous as texts, arising whenever more than one interpretation of a fact or phenomenon is possible. Also like texts, arguments are necessarily incomplete, leaving implicit much of the knowledge and reasoning from which they are constructed. Thus, fully grasping an argument may require as much close, inferential processing of its content as comprehending a text does. In this sense, argumentation is as much a form of reasoning as reading is, whether one is evaluating someone else's argument or constructing and justifying one's own.
Because reading, writing, arguing, and reasoning are so tightly intertwined in learning, it would be reasonable to expect that the skills and structures of argumentation would receive a great deal of attention in schools. Voss, however, says, "Most children, even by the end of middle school, are very ill-prepared to construct, understand, or assess arguments. They can't pick out the major points of an argument. Very few can write an argument-based paragraph. They don't know the components of arguments--premises, conclusions, counterexamples--because these aren't taught in school."
An inspection of elementary history texts conducted by a doctoral candidate working with Voss showed, for example, that the texts "tend to be written like narratives, and in the narratives there is not much causal explanation or justification --just, one thing happened, and it might have caused something else--but there's very little analysis of the events and motivations." In addition, this line of research showed, in studies that asked 5th-, 7th-, 9th-, and 11th-grade students to respond to questions requiring informal reasoning, that students with the highest general ability levels did better than those of average or lower ability. Surprisingly, though, says Voss, "Kids in 11th grade were not doing better than kids in 5th grade. In fact, the high-ability 5th graders were doing better than the mid- to low-ability 11th graders." This lack of improvement in informal reasoning from grade to grade is a consequence, in Voss's view, of the fact that schools do not teach the skills or components of argumentation.
Nevertheless, Voss points out, research has shown that children can generate effective arguments, outside school, in situations that are important to them. They can skillfully persuade parents to grant privileges or convince friends to concede a point of view, presumably because they have mustered the necessary information to bolster their position. "But ask those same kids to construct an argument in history, and they can't do it," Voss says. "If you add some instruction in argument along with the history content," he suggests, "that might change." Because Voss believes schools and texts should build on children's demonstrated ability to argue by explicitly teaching argumentation and causal analysis, he has conducted empirical investigations into the nature of informal reasoning in social science issues. In particular, he has examined the process of argument generation in social controversies such as those surrounding gun control, abortion, the death penalty, and testing for AIDS or for drug use. This work is based on earlier studies of the inferential processes by which people build mental representations of a problem in order to reason about it. The work on argument emphasizes the components of those inferential processes rather than the construction of mental representations.
For example, Voss says, "To understand something in social science, such as a historical event, learners need to analyze it, reconstruct it, put the pieces in order and in the right relation to each other. In order to do that, they need background knowledge, the skills that allow them to apply and connect that knowledge, and the ability to access information as they need it--in other words, to retrieve their background knowledge from memory." Argument construction is one mechanism for analyzing a social science concept or event, and in this case, as Voss explains, "How people generate, justify, evaluate and recall arguments has impact on their ability to think and communicate about topics in a subject area."
Voss began working with college students to examine these processes, asking them to generate as many pro and con arguments as they could in regard to various stated positions on controversial issues. Most people generated very few--a finding that Voss attributes to "output interference," or the fact that accessing memory for information in support of some arguments seems to inhibit further searchers for more information--but of those few, more tended to be in support of the students' personal views than in opposition to them. Even with the aid of cues (rather like the high-level prompts given in Chi's studies), subjects generated only one or two additional arguments. Thus, according to Voss, if a given study involved 40 people, they might collectively generate only about 15 pro and 15 con arguments; and a single individual might come up with only three of the 15 that supported his personal view and only one or two of the 15 that opposed it.
Moreover, Voss adds, if those same 40 people were asked to return 2 weeks later and repeat the same exercise--generating arguments for and against the same proposition--they would repeat only about one-third of their original arguments. All the other arguments offered in the second session would be new ones, which in Voss's view suggests that these "argument structures," or mental representations, "are not very stable and that the arguments that are repeated are the ones that the person holds as being stronger." Voss believes, "People tend to have a few arguments in mind that they associate with a given issue, and they probably generate the rest pretty much on the spot. The fact that they don't generate very many and that most of them are not stable suggests that they have relatively few ideas--not very much information--about the issue."
From these studies, Voss concludes about argument generation what Perfetti concludes about reading--that background knowledge plays a greater role than skill, as important as skill has also been shown to be. The fact that children can generate persuasive arguments with parents or friends, even without being taught the structure and components of argument, suggests that, as in reading, knowledge can compensate somewhat for lack of skill. This suggestion is further supported by the finding that people tend to generate most of their arguments in support of their own beliefs, which Voss believes is because they have more background knowledge that upholds their opinions.
Nevertheless, Voss maintains, it is necessary for schools to provide instruction in the skills of argument, and to encourage students to practice them frequently, so that the content already covered by the curriculum can combine with skill to support true reasoning. Voss's newer work, therefore, is a joint effort with NRCSL's John Levine to study ways in which group processes influence the development of informal reasoning skills, the kinds of training or experience that could facilitate reasoning in the classroom, and the extent to which social science instruction should place more emphasis on causal explanations and justifications.
-###-
[Argument and Changes in Thinking ]