I begin with a caveat: If you ever read a book on argumentation or take a course in argument, you’ll likely see more distinctions among the different types of evidence than what I’m about to present. I’ve condensed the many different types into three general categories: 1) examples 2) statistics and 3) expertise. I chose to discuss them this way for a couple of reasons: First, as I began writing this post, I became increasingly skeptical that finer distinctions would make the subject of evidence any clearer for my intended reader. In fact, as many textbook authors well know, decisions about how to treat material is often as much about marketing strategy as it is about teaching strategy. Second, when we start applying tests of evidence in the effort to avoid being seduced by fallacies, we’ll find the various manifestations of evidence in each of these three categories similar types of scrutiny.

Examples

Examples are presentations or re-tellings of the world around us. They include historical and hypothetical instances, stories and narratives. Historical examples (something that has already happened) include pointing to past accomplishments. When Bartiromo asked Kasisch how he would cut the budget, he cited the example of his success as governor of Ohio: In Ohio, we reduced Medicaid funding for the poor from 10 percent to 2.5 percent, didn’t cut one benefit or didn’t take anybody off the rolls. Why? Because we’re innovators.” [4th Republican Debate] [Watch it] When asked if Sanders was “tough enough” on gun control issues, Clinton and O’Malley cited the example of Sanders’ vote to grant gun manufacturers immunity.

Rarely does a single example provide enough weight to tip the balance toward a stronger argument (relying on too few examples results in a fallacy of hasty generalization). Conversely, a single example is enough to disprove a sweeping assertion.

Assertion: The United States only supports other democracies around the world.

Evidence: The United States supports Saudi Arabia, a totalitarian regime.

Reasoning: Since the United States supports this totalitarian regime, it isn’t true to say the United States only supports democracies.

In this case, you can see that a single example proves the assertion false.

Hypothetical examples usually carry less weight because they are “what if” scenarios. They attempt to provide an example of what a candidate’s plan would look like in action. In the Republicans debate scheduled for 12/15/15, you can be certain each candidate will provide a hypothetical example of how their tax plan would affect a specific group of taxpayers.

Political candidates often employ what they hope their listeners will find a compelling narrative, but again, these don’t serve as evidence. They have much more rhetorical currency when used as ethical or emotional means of persuasion.

What to look for in examples:

Is an example being peddled as a fact? What proof is there that the example or instance actually took place? Do we have any proof other than the candidate’s claim? Is the example highly contextualized or linguistically manipulated to make it appear as a fact?

Does the example resonate with our own experience or is it counterintuitive? Does the example seem like something you have experienced or does it seem far-fetched? Trump has become infamous (at least to those outside his bubble of supporters) for using examples many people don’t find credible. When fact-checked, several of the examples cited in Carson’s autobiography seem unable to pass the truth test as well.

Does the example provide enough weight in terms of both quantity and quality to allow a generalization? Sometimes, examples are expressed as evidence in support of an assertion, but actually serve more in the capacity to illustrate or clarify. This is usually the case with narratives. Listen to the next Republican debate on 12/15/15 and see when a narrative is used to illustrate a position.

Statistics

Statistics are numerical representations of the world around us. They include polls, snapshots and trends. Polls will dominate the news cycle from now until Election Day (For an excellent source of political data look at Nate Silver’s fivethirtyeight.com). Snapshots are data about a particular moment in time and trends show the evolution of data over time. You’ll find both as the results of various polls about the popularity of candidate’s are reported. Some excellent sites for polling and other statistical data are at The Pew Research Center, The Gallup Organization and The Quinnipiac University Polls.

What to look for in statistics:

Who conducted the poll? Fivethirtyeight has graded over 300 polling organizations based on their accuracy, methodology and bias. Their grades range from “A” to “F.” Don’t settle for candidates saying simply “According to a recent poll…” See Trump’s use of flimsy poll data.

What exactly, were the questions asked? We saw the potential problem with questions of definition in the post on assertions. The answers people give in a poll will depend on the questions they’re asked.

Expertise

Expertise is the interpretation of recognized authorities about the world around us. Among these recognized authorities, we may find those with academic or professional credentials, the findings of commissions or special committees and white papers.

When Clinton cited Nobel prize winning economist Paul Krugman as someone who agreed with her plan to reign in the excesses of Wall Street, she was using expertise as evidence.

What to look for in expertise

Does the recognized authority have some expertise in the subject? After much deliberation I didn’t include experience in the category of expertise. Someone may have been doing something a long time, but doing it poorly. Also, some news programs routinely feature as “experts,” strategists from opposing camps who provide banter instead of insight. What qualifies someone as an “expert” on a subject other than the fact he or she has been reporting on it?

Is there any reason to believe the expertise is biased? Some organizations self identify as approaching subjects with a political agenda. On the right, think Cato Institute, Freedomworks and the Heritage Foundation. On the left, think Center for American Progress, Media Matters and MoveOn.

Some Final Thoughts About Evidence

I’ll leave you with three additional thoughts about the nature of evidence in general:

 

  • The confidence level of someone offering evidence has no bearing on its truthfulness. I’ve sat through any number of meetings where someone uttered a fact that was absolutely false. I can only surmise they do this because they don’t know it’s false (speaking from ignorance) or because they thought they wouldn’t be challenged (speaking with the intent to mislead). In the public sphere, fact checkers measure assertions and evidence against reality. Two of the more trusted sources are Politifact, with it’s “pants on fire meter” and The Washington Post’s Fact Checker.
  • The number of people who believe a piece of evidence is true has no bearing on its truthfulness. Those who point to the large number of people who agree with a fact as evidence of its truthfulness commit the fallacy of appeal to public opinion. The exception to this of course, is a poll, where the point is to find out what people believe or agree with.
  • Facts do not speak for themselves. They demand interpretation and contextualization. The same piece of evidence can be used to support different, even contradictory assertions. For example, one could take the fact that Clinton has changed her position on several issues (gun control, the vote authorizing the use of force in Iraq, her assessment of the Free Trade Agreement, just to name three) as evidence in support of the assertion that “Clinton takes positions to coincide with current public opinion,” or the assertion that “Clinton is open to admitting her mistakes.” Don’t assume everyone interprets a fact the same way you do and at look at all the evidence from the other’s POV.

 

Coming up next: How reasoning provides the bridge between evidence and assertions.