Previous Month | RSS/XML | Current | Next Month

WEBLOG

January 31st, 2018 (Permalink)

A Strunk & White Puzzle

The seventeenth rule in William Strunk and E. B. White's famous stylebook The Elements of Style is: "Omit needless words." This is good advice to writers, but it is even more important with respect to letters. Can you omit needless letters from the following to produce a complete sentence?

ANECEODMLPELESTESSLEENTTETNERCES

Solution

January 30th, 2018 (Permalink)

New Book: Deciding What’s True

This book by Lucas Graves is not all that new, having been published in 2016, but it's new to me. For once, I have no complaints about the title or the subtitle, which is: "The Rise of Political Fact-Checking in American Journalism". So, this book appears to be a history and, perhaps, critique of political fact-checking.

Prior to the last decade, the word "fact-checking" usually referred to a process that was largely invisible to the average reader. The "fact checkers" worked for magazines and publishers, checking factual claims in articles and books prior to publication. This was usually a low-paid, entry-level job for young people, especially women.1

In contrast to this private fact-checking, Graves is concerned with the public political fact-checking that has developed in the last ten to fifteen years. Starting with Snopes, mostly websites began to check the claims made by politicians and pundits for accuracy. Snopes started as an urban legends site, having been created by a married couple of folklorists, but eventually expanded into checking claims made in chain emails2―a precursor to the "fake news" spread on Facebook and Twitter. Annenberg Political Fact Check seems to have been the first dedicated, full-fledged political fact-checking site, and is still the best, in my opinion.

Oddly enough, pre-publication fact-checking has been in serious decline during the same period as the rise of political fact-checking. As low-status employees, fact-checkers were often the first to go when publications began tightening their belts due to revenue lost to declining advertising. I wonder whether this change has any bearing on the simultaneous rise of "fake news" in the last several years that everyone, including the president, now complains so much about. How many false, if not "fake", news stories have made it into print, pixels, or on the air because there were no internal fact-checkers left to stop them? I don't know whether Graves' book addresses this interesting and perhaps important issue: I doubt it, but I hope so.

So, you might ask: how does this book relate to this website? Good question! There are two ways that reasoning can lead us into falsehood:

  1. The reasoning itself is incorrect, that is, fallacious or at least uncogent. This, of course, is what this website is mainly about.
  2. The premisses we start out with are false. If you start off with false claims, even if you reason entirely correctly, it will be a miracle if you end up at a true conclusion. In other words, garbage in, garbage out.

For this reason, getting your premisses right is every bit as important as reasoning correctly. So, fact-checking is just as important as logic-checking.

Furthermore, there are important logical and philosophical issues surrounding fact-checking: for instance, just what is a fact anyway? And how, exactly, do you decide what's true? I don't know whether Graves' book addresses these issues, but they deserve addressing, and I may have a thing or two to say about them.

Notes:

  1. For a history of fact-checking at Time magazine, see: Merrill Fabry, "Here’s How the First Fact-Checkers Were Able to Do Their Jobs Before the Internet", Time, 8/24/2017. By the way, this article really doesn't tell you how fact-checkers were able to check facts before the internet, but research was not impossible prior to the web or even computers: it was just slower.
  2. Michelle Dean, "Snopes and the Search for Facts in a Post-Fact World", Wired, 9/20/2017. Everything you ever wanted to know about Snopes, plus much more that you didn't want to know! Also, another dumb title: how can you look for facts if there are no longer any facts?
January 24th, 2018 (Permalink)

What's New?

I've added a new fallacy to the files: The No-True-Scotsman Fallacy! As was true of the previous new fallacy*, this is not an entirely new fallacy in that it used to be appended to the entry for Redefinition as a subfallacy. Now, I've given the Scotsman his own entry, and added a new leaf to the Taxonomy for him. In the process, I've revised both entries and given the Scotsman his own example, which was missing in the previous treatment.

*What's New?, 12/1/2017

January 9th, 2018 (Permalink)

Quote Watch: Don't Be Fooled

I've just started reading a book entitled Don't Be Fooled: A Citizen's Guide to News and Information in the Digital Age, which certainly sounds useful. In its preface the author, John H. McManus, offers a short autobiography, which begins:

In this book I argue that information-providers should be transparent so you'll know when to power up your skepticism shields. Nelson Mandela once said "where you stand depends on where you sit."1

So, let's raise our skepticism shields. Did Nelson Mandela actually once say that? First of all, what does it even mean? I sit in my chair, but I don't stand in it. Clearly, it's supposed to have something to do with bias, since the preface in which it occurs is titled "A Confession of Bias". "Where you stand" can refer to the "positions" you take on issues, but what does "where you sit" mean? Here's how McManus interprets it:

Without being aware of it, we absorb biases from where we are situated in society. Our race, gender, religion, generation, geography, class and nationality each have a great deal to do with how we perceive the world.1

Okay, but why should all of that be called "where you sit", as opposed to "where you're from" or "who you are" or "where you lie down"? Anyway, the fact that the quote doesn't make much sense coming from Mandela should help raise those skepticism shields.

If you search for this quote on the internet, you'll find many sites that attribute it to Mandela, including some quotation sites. However, none of those occurrences that I have found include a citation to a source for the quote. Also, we have seen previously that many quote sites are not to be trusted2, as they often include fake quotes and misattributions. Finally, the earliest attribution of the quote to Mandela that I've found is from 2010, which certainly doesn't rule out that he said it, but it is further reason to doubt it. Moreover, prior to 2010, the quote has been called "Miles' Law" after Rufus Miles.

Who was Rufus Miles3? If you're like me, you've never heard of him before. One common phenomenon of quotesmanship is that things said by little-known people―such as Rufus Miles―are put into the mouths of famous people―such as Nelson Mandela. Ralph Keyes calls this the axiom that "famous quotes need famous mouths"4. Even better, of course, if the famous mouth belongs to a highly-admired person, such as Mandela. In that way, the quote is given an air of authority that it would otherwise lack. Most people would be skeptical of a quote attributed to someone they've never heard of, but not so if the same quote is put in the mouth of a famous wise man.

In an article from 1978, Miles himself claimed authorship of the quote5, and here is how he explains the "law":

…[T]he basic lesson of Miles' Law was that there was no such thing as pure objectivity in the arena of budgeting or public policy making in general. Every person has a function to perform and that assigned responsibility markedly influences one's judgment.6

Miles was a bureaucrat writing for the Public Administration Review, and his law is one of seven "maxims" of management7. It's clear from Miles' discussion and examples that what he meant by "where you stand depends on where you sit" is that the "positions" that a bureaucrat takes on issues depend upon where in the bureaucracy he is positioned. So, where he "stands" depends upon where he "sits", that is, where his office and his chair is, what building he works in, what department or agency he works for.

Of course, the fact that Miles claimed authorship of the "law" doesn't necessarily mean that he did author it. Did Nelson Mandela at some time in his long life say it? Possibly; I certainly can't prove that he didn't8. However, even if he did say it, it's likely that he was repeating something he had heard. It is far more likely that Miles came up with the saying than Mandela did, and the clincher for me is the fact that the quote makes more sense in its bureaucratic setting than as a general maxim about bias.

I've just begun to read McManus' book, and it may well be a highly useful and accurate one despite its inauspicious beginning. McManus either did not raise his "skepticism shields" when confronted with the alleged Mandela quote, or he's not much of a researcher―I hope the former. It's so easy to discover that the attribution of the quote to Mandela is at least doubtful that I suspect that McManus just didn't bother to check it. Instead of the purported Mandela quote, McManus should have kept in mind the old journalistic saying: "If your mother says she loves you, check it out."9

Notes:

  1. John H. McManus, Don't Be Fooled: A Citizen's Guide to News and Information in the Digital Age (2012), "A Confession of Bias"
  2. See, for instance: Who is Adolph Hitler and why does he keep saying these terrible things?, 9/11/2017
  3. You can read his short obituary here: "Rufus Miles Jr., 85, Aide to 3 Presidents", The New York Times, 4/15/1996
  4. Ralph Keyes, "Nice Guys Finish Last Seventh": False Phrases, Spurious Sayings, and Familiar Misquotations (1993), p. 20
  5. Rufus E. Miles, Jr., "The Origin and Meaning of Miles' Law", Public Administration Review, Vol. 38, No. 5 (Sep.-Oct., 1978), pp. 399-403
  6. Miles, p. 400
  7. Miles, p. 403
  8. Moreover, the burden of proof is not on me to disprove it. The burden is on McManus or anyone else who claims that Mandela said it to offer some evidence that he did. So far, there is no evidence he did.
  9. Chip Scanlan, "If Your Mother Says She Loves You: A Reporter's Cautionary Tale", Poynter, 4/17/2003
January 5th, 2018 (Permalink)

Poll Watch: Big Nothin'

A lot of public opinion poll reporting is fake news in the sense that there's no news in it1. More precisely, it might be called "manufactured news" since the "news" is the poll itself. If the numbers have changed since the previous poll, that's news! If they haven't, that's news too! A current example has the following headline:

Trump weekly job approval up to 39 percent, highest in months

The article under this headline2 is quite short, though not as short as it should be. If you're an experienced poll watcher, I suggest reading the whole article, which takes less than a minute. Then, you can compare your observations to my comments, below:

Excerpt Comment
The approval rating for the week was an improvement from Trump's low of 35 percent, seen just weeks before.
Four percentage points? Is that significant? What is the margin of error for this poll? That's not reported!
The last time Trump's weekly approval was at 39 percent was in July, according to Gallup. The president…had an overall approval average of 39 percent for 2017.
So, another headline for this report could have been: Trump Job Approval Rating Holds Steady
The president's latest weekly approval rating in December came after he scored his first legislative victory with congressional Republicans on tax reform.
So that probably caused the increase….
However, the Republican-backed tax plan does not appear to be widely popular with the American people. An NBC News–Wall Street Journal poll last month found that only 24 percent of Americans believed the tax plan was a good idea.
…um, never mind.

According to Gallup3, the margin of error (MoE) for this poll is ±3 percentage points. However, the previous week's approval rating was 37%, which is within the MoE. Furthermore, as recently as the middle of November, the rating was at 38%. So, in about six weeks the rating improved one whole percentage point.4

You can't blame the reporters who write this stuff: they've got a job to do, and they can't let the fact that there's no news stop them.

Notes:

  1. For a "Poll Watch" example from last year, see: No Margin for Error, 5/4/2017
  2. Julia Manchester, "Poll: Trump weekly job approval up to 39 percent, highest in months", The Hill, 1/3/2018
  3. "Trump Job Approval (Weekly)", Gallup, accessed: 1/5/2018.
  4. See: Margin of Error Errors, How to Read a Poll.

Solution to a Strunk & White Puzzle: Remove the needless letters "NEEDLESS LETTERS" to get "A COMPLETE SENTENCE".

Previous Month | RSS/XML | Current | Next Month