2 Chapter 2 (Introducing Research)

Joining a Conversation

Typically, when students are taught about citing sources, it is in the context of the need to avoid plagiarism. While that is a valuable and worthwhile goal in its own right, it shifts the focus past one of the original motives for source citation. The goal of referencing sources was originally to situate thoughts in a conversation and to provide support for ideas. If I learned about ethics from Kant, then I cite Kant so that people would know whose understanding shaped my thinking. More than that, if they liked what I had to say, they could read more from Kant to explore those ideas. Of course, if they disliked what I had to say, I could also refer them to Kant’s arguments and use them to back up my own thinking.

For example, you should not take my word for it that oranges contain Vitamin C. I could not cut up an orange and extract the Vitamin C, and I’m only vaguely aware of its chemical formula. However, you don’t have to. The FDA and the CDC both support this idea, and they can provide the documentation. For that matter, I can also find formal articles that provide more information. One of the purposes of a college education is to introduce students to the larger body of knowledge that exists. For example, a student studying marketing is in no small part trying to gain access to the information that others have learned—over time—about what makes for effective marketing techniques. Chemistry students are not required to derive the periodic table on their own once every four years.

In short, college classes (and college essays) are often about joining a broader conversation on a subject. Learning, in general, is about opening one’s mind to the idea that the person doing the learning is not the beginning nor the end of all knowledge.

Remember that most college-level assignments often exist so that a teacher can evaluate a student’s knowledge. This means that displaying more of that knowledge and explaining more of the reasoning that goes into a claim typically does more to fulfill the goals of an assignment. The difference between an essay and a multiple choice test is that an essay typically gives students more room to demonstrate a thought process in action. It is a way of having students “show their work,” and so essays that jump to the end without that work are setting themselves up for failure.

Learning, Not Listing

Aristotle once claimed “It is the mark of an educated mind to be able to entertain a thought without accepting it.” Many students are familiar with the idea of search. The internet makes it tremendously easy to search for information—and misinformation. It takes a few seconds to find millions of results on almost any topic. However, that is not research. Where, after all, is the re in all of that? The Cambridge Dictionary offers the following definition:

Research (verb): to study a subject in detail, especially in order to discover new information or reach a new understanding.

Nothing about that implies a casual effort to type a couple of words into search engine and assuming that a result on the first screen is probably good enough. Nor does that imply that a weighted or biased search question, like “why is animal testing bad” will get worthwhile results.  Instead, research requires that the researcher searches, learns a little, and then searches again. Additionally, the level of detail matters. Research often involves knowing enough to understand the deeper levels of the subject.

For example, if someone is researching the efficacy of animal testing, they might encounter a claim that mice share a certain percentage of their DNA with human beings. Even this is problematic, because measuring DNA by percentage isn’t as simple as it sounds. However, estimates range from 85% to 97.5%, with the latter number being the one that refers to the active or “working” DNA. Unfortunately, the casual reader still knows nothing of value about using mice for human research. Why? Because the casual reader doesn’t know if the testing being done involves the 97.5% or the 2.5%, or even if the test is one where it can be separated.

To put it bluntly, Abraham Lincoln once had a trip to the theater that started 97.5% the same as his other trips to the theater.

In order for casual readers to make sense of this single factoid, they need to know more about DNA and about the nature of the tests being performed on the animals. They probably need to understand biology at least a little. They certainly need to understand math at a high enough level to understand basic statistics. All of this, of course, assumes that the student has also decided that the source itself is worthy of trust. In other words, an activist website found through a search engine that proclaims “mice are almost identical to humans” or “mice lack 300 million base pairs that humans have”. Neither source is lying. Just neither source helps the reader understand what is being talked about (if the sources themselves even understand).

Before a student can write a decent paper, the student needs to have decent information. Finding that information requires research, not search. Often, student writers (and other rhetors) mistakenly begin with a presupposed position that they then try to force into the confines of their rhetoric.  Argumentation requires an investigation into an issue before any claim is proffered for discussion.  The ‘thesis statement’ comes last; in many ways it is the product of extensive investigation and learning. A student should be equally open to and skeptical of all sources.

Skepticism in Research

Skepticism is not doubting everyone who disagrees with you. True skepticism is doubting all claims equally and requiring every claim to be held to the same burden of proof (not just the claims we disagree with). By far, the biggest misconception novice writers struggle with is the idea that it is okay to use a low-quality source (like a blog, or a news article, or an activist organization) because they got “just facts” from that source.

The assumption seems to be that all presentations of fact are equally presented, or that sources don’t lie. However, even leaving aside that many times people do lie in their own interests, which facts are presented and how they are presented changes immensely. There’s no such thing as “just facts.” The presentation of facts matters, as does how they are gathered. Source evaluation is a fundamental aspect of advanced academic writing.

“Lies” of Omission and Inclusion: One of the simplest ways to misrepresent information is simply to exclude material that could weaken the stance that is favored by the author. This tactic is frequently called stacking the deck, and it is obviously dishonest. However, there is a related problem known as observational bias, wherein the author might not have a single negative intention whatsoever. Instead, xe simply only pays attention to the evidence that supports xir cause, because it’s what’s relevant to xir.

• Arguments in favor of nuclear energy as a “clean” fuel source frequently leave out the problem of what to do with the spent fuel rods (i.e. radioactive waste). Similarly, arguments against nuclear energy frequently count only dramatic failures of older plants and not the safe operation of numerous modern plants; another version is to highlight the health risks of nuclear energy without providing the context of health risks caused by equivalent fuel sources (e.g. coal or natural gas).
• Those who rely on personal observation in support of the idea that Zoomers are lazy might count only the times they see younger people playing games or relaxing, ignoring the number of times they see people that same age working jobs or—more accurately—the times they don’t see people that age because they are too busy helping around the house or doing homework.

A source that simply lists ideas without providing evidence or justifying how the evidence supports its conclusions is likely not a source that meets the rigor needed for an academic argument. While later chapters will go into the subject in greater detail, these guidelines suggest that in general, news media are not ideal sources. Neither are activist webpages, nor are blogs or government outlets. As later chapters will explore, all of these “sources” are not in fact sources of information. Inevitably, these documents to not create information, they simply report it. Instead, finding the original studies (performed by experts, typically controlled for bias, and reviewed by other experts before being published) is a much better alternative.

Examining Sources Using the Toulmin Model

On most issues, contradictory evidence exists and the researcher must review the options in a way that establishes one piece of evidence as more verifiable, or as otherwise preferable, to the other.  In essence, researchers must be able to compare arguments to one another.

Stephen Toulmin introduced a model of analyzing arguments that broke arguments down into three essential components and three additional factors. His model provides a widely-used and accessible means of both studying and drafting arguments.

At a basic level, the Toulmin model breaks arguments down into three parts: claim, data, and warrant. The claim is whatever the writer is trying to prove (e.g. ‘it is cold in this room’ or ‘you should stop smoking’). The data are whatever support is used to build up the claim (e.g. ‘the thermostat says it is sixty-two degrees Fahrenheit’ or ‘smoking is linked to heart disease and to lung cancer’). Finally, the warrant represents an assumption or a shared understanding that allows the data to support the claim (e.g. ‘sixty-two degrees is cold’ or ‘you should make personal decisions in the interest of your long-term health’). Note, however, that each of these three parts can become its own claim—someone might argue over how cold is too cold for a room or whether or not the studies on smoking are valid.

The Toulmin model can be complicated with three other components, as well: backing, rebuttals, and qualifiers. Backing represents support of the data (e.g. ‘the thermostat has always been reliable in the past’ or ‘these studies have been replicated dozens of times with many different populations’). Rebuttals, on the other hand, admit limits to the argument (e.g. ‘unless the thermostat is broken’ or ‘if you care about your long-term health’). Finally, qualifiers indicate how certain someone is about the argument (e.g. ‘it is definitely too cold in here’ is different than ‘it might be too cold in here’; likewise, ‘you might want to stop smoking’ is a lot less forceful than ‘you absolutely should stop smoking’).

At a minimum, an argument (either one made by the student or by a source being evaluated) should have all three of the primary components, even if they are incorporated together. However, most developed arguments (even short answers on tests or simple blog posts) should have all six elements in place. If they are missing, it is up to the reader to go looking for what is missing and to try to figure out why it might have been left out.

Here is an example of an underdeveloped argument that is simply phrased like an absolute claim of fact. It is a poor argument, in that it offers none of the rationale behind what it says—it just insists that it is correct:

“Other countries hate the United States for a reason.”

What other countries? What reason? Is it just one reason, or is it one reason per country?

By contrast, here is an argument that has at least some minimal development:

“In the eyes of many (Qualifier), the United States has earned the hatred of other countries (Claim). The U.S. involvement in Iranian politics alone has earned the country criticism (Data). By helping to overthrow a democratically elected leader in favor of a monarch in 1953, the U.S. acted in a manner that seemed hypocritical and self-interested (Backing). While many countries do act in favor of their own interests (Rebuttal), the U.S. publicly championing democracy while covertly acting against it serves to justify criticism of the country (Warrant).”

Is there room to disagree with this argument? Yes. However, this argument provides its rationale, it offers at least some sort of evidence for its claims, and it provides a place to begin engagement. A researcher who wishes to know more about this argument can go looking into the history of U.S.-Iran relations, for example.

When reviewing a source, or making their own arguments, researchers should consider the following questions. Is there evidence that can be verified and examined by others (in the same spirit as the scientific method)? More specifically:

• What is the claim?
• What data backs up this claim?
• What assumptions do I have to make to consider this evidence to be adequate support?

The various pieces of data which support claims in the Toulmin model are often called into question.  Studies are refuted, statistics countered with rival numbers, and their applicability to the claim in question is often murky.  Evidence—whether offered as matters of fact or as subjective considerations—does not exist in a vacuum.  Data are themselves claims.  If the supporting data are accepted as true, the argument has a generally accepted conclusion.  Such pieces of ‘evidence’ are contentions.

Although Toulmin distinguishes between qualifiers and rebuttal conditions, such a distinction is difficult to maintain in practice.  The important consideration—the one acknowledged by both terms—is that unconditional or absolute claims are difficult to support.  Specific fields have their own ways of hedging their bets.  Science has its error bar (Sagan) and the terminology of probability.  Statistics and polling have a margin of error.  Ethnography has its confrontation of personal bias.  When a rhetor expresses the limitations of a given claim, when the unconditional becomes conditional, claims become more than categorical propositions or thesis statements.  They become arguments.

Example 1: Here is a minimalistic overview of one claim on the topic of traffic cameras.

• Claim = Traffic cameras increase minor accidents
• Evidence = David Kidwell and Alex Richards of The Chicago Tribune performed a study that was later cited by ABC News.
• Assumptions = This study was conducted honestly and reasonably represents the reality of accidents around these cameras (i.e. I can trust the agenda and the methods of the Chicago Tribune staff).

Example 2: And here is a second claim on the same subject.

• Claim = The types of accidents by traffic cameras tend to be less severe
• Evidence = The Insurance Institute for Highway Safety examined national trends and compared medical reports, police reports, and various bills, posting the results on their website.
• Assumptions = If the IIHS has a bias, it would be toward fewer accidents, or at least less severe accidents (because this means they have to pay less money out).

As an Essay Fragment: According to some, traffic cameras actually increase accidents. A study conducted by the Chicago Tribune found that rear-end collisions increased when traffic cameras were installed, meaning that they make things worse, not better (Kidwell and Richards). However, the Insurance Institute for Highway Safety points out that while there are sometimes increases in minor collisions, the number of crashes resulting in injuries actually decreases.

• “According to…not better (Kidwell and Richards).” This uses a parenthetical source citation to provide a “link” to the evidence and to invite readers to examine both the data and the warrants.
• “However, the…decreases.” This uses a signal statement to introduce the source of the evidence first, often because the source has so much credibility the author is hoping to impress the reader.

Note that this is not a particularly powerful fragment–it is simply the minimum level of rigor that a student should offer (or look for) in an academic essay or in an academic source.

Academic arguments typically make concessions.  These concessions help define the scope of the argument and the range of the inquiry.  In Section 1, I mentioned a relatively straightforward value claim: “Plan X is bad.”  Argumentation engages such value claims and defines their scope and limits.  Who is plan X bad for?  By what standards?  Why then is anyone in favor of plan X?  A more practical approach could be “If you favor Y, then Plan X is bad.”  This is a concession, of sorts—Plan X is only bad if you favor Y.  The argument admits that if you do not, then Plan X might not be all that bad, after all.

Such a concession, worded in such a way, has added merit.  It functions as what Aristotle would have called an artistic proof, although maybe not an enthymeme.  It establishes a bond between the rhetor and the audience through the shared favoring of Y; it nurtures consubtantiality—the basis of what Burke calls identification.  Clearly, concessions can be made in a way that both prevents some counterarguments from applying and still furthers a rhetorical point.

Such phrasing is practical, and only truly cynical interrogators would consider it sinister.  An inversion of this approach is possible.  “Unless you favor Y, Plan X is bad.”  So long as Y is sufficiently negative in the minds of the audience, the rhetor loses no actual impact here.  Here, connecting Y to X might require substantiation on the part of the rhetor, because the concession has become, itself, a justification of why X is bad (it is related to or involves Y).  The additional rhetorical power—gained through positive and negative associations—often compensates for such additional effort.