Woo!
[MAJOR UPDATE: I have changed “Woo” to “Pitch” everywhere on the website and on this post due to extensive feedback from everyone. Thanks!]
I’m adding rhetorical-device/common-argument/argument-fallacy tags to the expert quotes on TakeOnIt and calling them “pitches”.
The list of pitches so far is here.
Arguments have common patterns. The most notorious of these are rhetorical devices and argument fallacies. While these techniques are obviously not new and are published on several sites on the internet, they are woefully under appreciated by most people. I contend that this is partly because:
Argument fallacies and rhetorical devices can be too general. Most of their real-world usage occurs in a larger number of specialized forms. These specialized forms are often unlabeled yet are intuitively recognized and prey on our cognitive biases. It takes a lot of cognitive energy to consciously connect the general form(s) to the specialized form.
The sites about argument fallacies and rhetorical devices are not integrated with debate sites. A google for argument fallacies will give you pages with stagnant lists of fallacies where each one has perhaps a couple of historical or hypothetical applications of the fallacy. Why can’t I see every debate where some expert or influential person used that fallacy, and why can’t I see every fallacy used in a debate?
To solve these problems, I’m introducing the concept of a “pitch”. Any quote from an expert or influential person on TakeOnIt can now be tagged with a pitch. A pitch is a label for a commonly used argument or strategy to persuade. You can think of pitches as the “tv tropes of argumentation”. Here’s some examples:
“The Consensus Pitch”
“The Patriot Pitch”
“The Convert Pitch”
Pitches encompass both argument fallacies and rhetorical devices. However, they allow for greater specialization. For example, there is the “The Evil Corporation Pitch”. On a more minor note, I personally think the names should be simple and ideally guessable from the name alone (e.g. maybe it’s just me, but “Post hoc ergo propter hoc” feels like it has some Web 2.0 marketing issues).
Eliezer’s “Conversation Halters” and Robin Hanson’s “Contrarian Excuses” are good candidates for pitches. (My impression is the “halters” and “excuses” listed are perhaps too specialized for pitches, but in any case at minimum provide fertile material for pitches.)
I only implemented this feature over the last few days and before developing the concept further I’d like to get some feedback.
Why the name “woo”? The fact that this is likely to occur in the same context as “woo” as in “woo-woo” makes me uneasy; it seems likely someone not paying attention might actually get them confused. (Well, one is a count noun and one is a mass noun, but...)
Also, can you state the definition more explicitly? It’s not clear from the above whether you’re considering valid patterns of argument to be woos or not.
Yeah, that’s the first thing I thought of too. Especially since I often hear “woo-woo” itself abbreviated to just “woo”. I think it should be renamed, especially since a lot of people interested in the structure of argument are likely to be active in the skeptic community and therefore potentially familiar with the term in its older sense.
The term is actually derived from the verb to “woo”.
The definition “A woo is a label for a commonly used argument or strategy to persuade” encompasses any commonly used and persuasive argument, including both valid and invalid arguments, or arguments that may or not be valid depending on how they’re used (such as the Consensus Woo).
I think however an attribute attached to each Woo of its intrinsic validity would be a good idea. That kind of data could then be used to rate experts according to how often they use bad arguments, and hence contribute to the calculation of Eliezer’s Correct Contrarian Cluster.
The term “woo” is already taken; you need to choose another.
Actually, “woo” was “taken” before the year 1050:
“to seek to persuade (a person, group, etc.)...”
http://dictionary.reference.com/browse/woo
The fact that my usage is consistent with that I think is quite reasonable. Most people outside of the skeptics community aren’t actually familiar with the usage of the term “woo” that you’re referring too. Nonetheless, I obviously see your point here. It would be preferable if there was no usage-confusion—especially since the skeptics community is really important to me. Does anyone have an alternative suggestion?
“Fallacy” and “Rhetoric” are both more neutral and broader than “woo”. I’d even put a label on that says “Alleged fallacy” rather than “Fallacy”. Otherwise it’s simply a matter of accusing people with no way to defend themselves.
This sort of thing broadens out very quickly into Issuepedia 2.0, if you allow replies, say. It would make more sense to think about it carefully than to do it ad-hoc (although not, of course, to try to implement all features at once).
Or “technique”, or possibly “argument”.
Careful incremental steps are the way to proceed. Let me explain the current step I’m taking. People generally visit TakeOnIt from a Google search to find out opinions about a particular issue that they searched for. Let’s say that they arrive at this page, to find out opinions on whether evolution is true:
http://www.takeonit.com/question/27.aspx
I’ve annotated some of these quotes—it’s just a start—with “woos”, or whatever we want to call them. Now let’s say none of these quotes were annotated. The result? A person can be persuaded without seeing the general patterns of how they got persuaded. This happened to a friend of mine. He got persuaded by a quote where I was like: “you can’t see why this argument is duping you?!”. It took me a while to explain the persuasion tactics used in the quote. It was basically a slow process of me identifying and communicating persuasion patterns.
That’s where labeling the quotes comes in. It allows a smaller community who’s familiar with these patterns of persuasion to pre-process those patterns for the larger community. Now, there’s an additional step we could take. We could categorize the various kinds of persuasion patterns. So X is an “alleged fallacy” while Y is “rhetoric” and so on. I actually suspect that more than one category per label is required and it’s a mistake to think these labels naturally fall into discrete mutually exclusive categories. However, this label categorization step is secondary. The highest order bit is to simply label the persuasion in the first place. That’s where most of the cognitive work is.
I considered allowing a truly broader notion than a “woo”, which was simply to allow any tag at all on a quote. However, I think restricting the tags to persuasive patterns gives good focus and avoids dilution.
Perhaps you have not realized how this blows up your whole site.
At present it is a neutral record of expert opinions. Who said what. On the record.
To label something as “woo”—or even the far more innocuous “persuasion pattern”—is not a neutral act. People disagree about what is woo or not woo. They disagree about whether persuasion patterns are being used. They disagree about what constitutes a logical fallacy, both in general and in the specific.
If someone claims that the Singularity is religious woo, do I get a chance to defend myself? How?
You have just taken a giant step from recording expert opinions to trying and provide a way for your audience to accuse and counteraccuse experts of being biased. This is not a trivial step. And allowing people to label things as “woo” does not seem like the best first step.
That’s not the same “woo” as BenAlbahari is referring to—he’s trying to impose a new term (with a different etymology) that seems to have some accidental overlap with “woo” as you seem to be using it here (which has more negative connotations). Which is a very very bad idea.
Someone claims the Singularity is a religious, theistic persuasion pattern that offers its believers a happy afterlife while others are left in the cold—to give an example of a typical and common accusation that people just make up, not based on any evidence, but because their brain completes the pattern for what they expect.
Do I get to defend myself? How?
Let’s make this conversation non-hypothetical. Here’s your expert page on TakeOnIt. I tagged a few of your quotes with some pitches:
http://www.takeonit.com/expert/693.aspx
I see. Well, I don’t object to the labels that I see. But you’re allowing anyone to edit the pitch list. What happens in case of an edit war?
You get a cacophony.
Seriously however, I see this as highly comparable to editing a controversial Wikipedia page, such as a page on George Bush or Climate Change. Ultimately the moderators get the last say, but you make the edit history transparent. I’m happy for anyone with enough rep points on Less Wrong to be a moderator on TakeOnIt. To be honest, at this point, my hunch is that any hypothetical answer I have to this question will be overshadowed by what I discover happens in practice.
Matching a pattern is evidence.
Maintaining a Neutral Point of View (NPOV) is very important. The related concept is No Original Research (NOR). To express your contention in Wikipedia terms, you’re concerned that choosing “persuasion pattern labels” violates NOR, which in turn violates NPOV.
My thoughts:
Annotations are precisely that; they’re purely suggestions about the meaning of the quotes. They don’t actually alter the quotes, so the downside is bounded. To make a strange thought experiment, consider what would happen if the Chinese Government purely inserted tags on web content, rather than filtered web content. It’s not as damaging. You could perhaps even argue it could be less biased than purely unfiltered content, because it would expose the Chinese Government’s agenda.
Selecting “persuasion pattern labels” is within the acceptable bounds of NOR. Most of the labels have very specific meanings. The important question is whether a decent sized community can reach consensus on the assigning of labels. Let’s take the Less Wrong community. I would expect in most cases we’d rapidly reach consensus. Sure, there would be vigorous debate at times, but that’s no different than for Wikipedia. There’s always going to be people who will cry NPOV foul; that’s unavoidable. They can make the Conservopedia version of TakeOnIt and have all the tags have a religious focus rather than a logical one.
There needs to be guidelines for Persuasion Pattern Labels. The existing set of labels reek of the brainstorming phase. Some will be removed and I’ve obviously gone overboard with pejorative language. When these are cleaned up, I doubt there will be a serious issue with anyone labeling the Singularity with the “Religious” label. Sure, there will be a few poorly chosen labels, but so long as there’s many more well chosen than poorly chosen labels, it’s worth labeling.
P.S. Some people have suggested the term “pitch” instead of “woo”. This certainly seems to solve some of the complaints people have had about the name.
“Pitch” is much better. I had no idea what the “woo” link meant when I first saw it.
“Persuasion pattern” is perfect. Use that!
This fact is irrelevant. Only what people read in it these days matters. (History might affect present perception, but this is an indirect effect that is only visible through the effect on present perception.)
That is still present usage. I’d bet real money on finding more news headlines with “woo” in the sense of courting than in the other, more recent acceptation.
I do think the term will not serve Ben, but more due to context.
Maybe “trope”, by your analogy to TVTropes (which did help me understand what you were getting at in the first place). Outside the context of the site, where the term might have other meanings, they could be described more explicitly as something like “persuasion tropes”. (I was going to say “rhetorical tropes”, but apparently that already means something.)
I agree “persuasion tropes” captures the idea really well. I guess I’m after a term that is closer to “persuasion” than “trope”, and is simple (a single word) and catchy. “Appeal” is a word I played with, which I liked because it was already used in the names of many argument fallacies, but it seemed to me to lack a Web 2.0 sound (think Tweet, Digg, Trope).
Etymology of a word doesn’t get to decide the connotations elicited by it in the reader.
ata asked me “why the name” so I answered. I of course agree with you that connotations are more important than etymology. Frankly however, I was hoping for deeper feedback. If connotations were the only problem I was dealing with here, then I’d be very satisfied.
I get the impression you underestimate the importance of such trivial (to fix) defects.
The “why” questions may hide plenty of depth. Why did you choose this label? If you answer with etymology, then etymology must be for you the most important not-obvious-to-the-reader consideration about this decision, singling it out from all the other reasonable options. Otherwise, why would you give this particular detail of the process of reaching the decision? (Most likely, because of rationalization-causing bias, but then it’s not the real answer to the “why” question.)
I concede. The name is changed to “pitch”. HT loqi who suggested the name + everyone else who gave feedback.
I’m open to your suggestion that I’ve underestimated its importance. What’s your alternative suggestion then? (I elaborated below in my reply to Jack on the issues with naming.)
It’s just a really annoying sounding word. Also, neologisms should be avoided whenever possible.
Yes… the presence of unnecessary jargon is a truly loathsome feature of many classes of developed knowledge… should be avoided at all costs…
At one point I considered having any arbitrary tag for a quote. However, this was too open-ended. I thought it made sense to constrain the meaning of the tags to the tactics used to persuade. I then started thinking about categories of such tactics, and realized that the instances of persuasive tactics didn’t neatly neatly fall into categories. I found many tactics weren’t clearly classified as an argument or a rhetorical device, but somewhere in-between. Furthermore, I realized: what value is there in even deliberating over that choice? It seemed sufficient to simply have a term that captured the general case: a persuasive tactic. Now, I could have chosen the term “argument” but then some people will complain that they’re not all arguments. That’s how the new term came about.
Afaict a woo is sometimes a logical fallacy, sometimes an argument form, sometimes the use of particular word. It is kinda of confusing that you throw in perfectly good arguments right next the worst logical fallacies. Why not distinguish fallacies, argument schema and other rhetorical devices? Then you’d have the added benefit of teaching people basic logic.
There seems to be an emphasis here on “bad” (logically invalid/irrelevant) forms of persuasion, but it might be also be nice to have “good” woos that should lend credence to an argument. This would also reduce the number of apparent epistemological assumptions made by the site (e.g., “relies on infinite set theory” may be more of a problem for some than others).
A few off the top of my head:
Disjunctive: Argument draws support from multiple independent sources.
Fragile: Argument makes very specific claims and/or is very vulnerable to new evidence.
Uncritical Consensus: Argument claims that no reasonable effort has been made to refute its core claims (e.g., citing ciphergoth’s recent cryonics survey).
FYI, I also think that “woo” isn’t a great word, and much prefer “trope” or even “pattern”. Maybe “pitch”?
Repeating it here: “Persuasion pattern”
Seconding pitch, if you actually want to use a new word rather than an existing one.
“pitch”… I like it. Anyone else? If there’s a consensus here, I’ll change it.
To me, “pitch” immediately suggests music/acoustics.
(Though, if you know the meaning of my pseudonym, you won’t find that surprising.)
*search engines*
Esperanto?
Yes, although any German speaker will also recognize the meaning.
“Pitch” is much better. I had no idea what the “woo” link meant when I first saw it.
I think you should be less interested in the positive consensus on “pitch” and more interested in the negative consensus on “woo”. I suspect you’re going to have to make many more decisions like this in the future.
I thought that TakeOnIt was documenting what experts believe, and only documents the arguments they make because it is evidence of their beliefs. It seems that picking apart the snippets of arguments quoted on TakeOnIt is an attack on strawmen.
These woos would be a good feature for a site dedicated to presenting the best arguments an expert makes. I am not saying the TakeOnIt should endeavor to present the best arguments (its documentation of expert opinion is valuable), but that since it does not, it should not presume to critique the arguments it happens to document in support of its primary purpose.
Assuming a descriptor is accurate and unbiased, the object of description remains unchanged. The accurate labeling of arguments does not constitute a critique, and the purpose of the website remains the same. (While an accurate labeling may be interpreted as a critique, that interpretation lies on top of its first intention.) It could make sense to argue for an inescapable bias of descriptors (see EY and BA below), but that is a different matter.
The problem is that the object of description is changed by merely being quoted on TakeOnIt. It is taken out of its context (possibly including supporting arguments), and presented as if it were the response to a question. (Fortunately, it does link to the source of the quote, which may provide context, but this relies on users actually following the link.) There is potential for bias in selecting what argument to quote as evidence that an expert holds a particular opinion, especially since a strongly stated conclusion is better for purposes of demonstrating that the expert has a given opinion, than the detailed arguments and evidence that support that conclusion. An accurate label on the quote could inaccurately reflect on the expert and the validity of their opinion.
TakeOnIt is not a good resource for determining what arguments and evidence support an expert’s opinion. Documenting woo types is an inaccurate signal that it should be used as such a resource. To avoid tricking people by quoting misleading rhetoric, it may be best to explicitly only quote statements of conclusion and exclude the arguments.
On quotes:
Quotes are by their nature vulnerable to being taken out of context. Any author who quotes an expert, or attributes a viewpoint to an expert, makes some effort to do so accurately and fairly. If they overly misrepresent people, they risk damaging their reputation. The same holds true for TakeOnIt. I should note that TakeOnIt, unlike other quote websites, always provides a source to help guard against that.
On deconstructing arguments:
An issue can be broken down into several sub-issues, where each issue has its own dedicated question on TakeOnIt. That way, you can see not only a summary of the expert’s opinion on a particular issue, but also their opinions on more specific issues that lead them to their main conclusion. For example, here’s Roy Spencer, a climate skeptic’s opinions:
http://www.takeonit.com/expert/238.aspx
You can see not only his opinion on whether AGW is true, but also his specific opinions that lead him to that (in this case, negative feedback loops).
On collating an expert’s arguments:
I was toying with the idea of displaying links to the quotes to sub-issues directly below the quotes for high-level issues. When I think about it, I’m not sure why I haven’t coded up that feature...
That seems like a reasonable way to analyze an expert’s opinion.
It would be good to have the list of sub issues, saying whether the expert agrees or disagrees (or mostly agrees/disagrees, etc) and link to their quote on that issue.
I’ve implemented it. Check it out:
http://www.takeonit.com/question/5.aspx
Nice
OK, I’ll add this feature.
If this interpretation of TakeOnIt is correct, I can’t see the website being very useful. Pure conclusions count little or nothing for me—it’s all about the argument.
It is valuable for evaluating experts based on their ability to reach accurate conclusions. See The Correct Contrarian Cluster.
And you can, of course, follow the citation to link to the full argument.
TakeOnIt has a separate construct for specific arguments supporting a position. Look at the arguments tab on the Global Warming debate or Cryonics debate:
http://www.takeonit.com/question/5.aspx
http://www.takeonit.com/question/318.aspx
The arguments tab works by connecting questions together via logical implications, such that the answer to one question implies the answer to another.
In comparison, pitches are just a way to mark up quotes such that well known persuasion patterns can be identified.
The very first thing I thought of was “how can he possible do this without people shooting him down ‘why did you label my favorite person a woo?!?’?”
Only after a while did I realize you didn’t want the two terms to be related. Don’t expect that from anyone emotionally invested in defending their hero.
Those of us reading this are abnormal. But we may not be sufficiently different from the people reading TakeOnIt to justify keeping Woo “Woo” despite its obvious “implications”.
I know there is a certain amount of poetry to it but I know a 1 year old who refers to ‘woos’ quite frequently and the name, particularly in the context of ‘undesirable persuasive crap’ just doesn’t sound right.
Even apart from sounding like potty talk I had associated the term ‘woo’ with a somewhat different. I admittedly haven’t been exposed to a culture that used the term woo frequently but it seems to fit better as “that is woo” than “that is a woo” and “those are woos”. This is perhaps because I associated it with the term ‘bullshit’. A different technical meaning but a similar grammatical class.
(I don’t actually object to using technical terms for specific kinds of bad thinking and communication strategies. Woo has its place somewhere.)
Upvoted, but is anyone else confused about what the Oprah Clap Pitch is?
(Edit 2010/03/17: It was eventually renamed The Heartfelt Pitch.)
If others find it confusing, we should rename it, change it, or get rid of it. If a pitch is not obvious to Less Wrong readers who are incredibly familiar with argument fallacies etc., it’s not going to be obvious to anyone.
It maps to the Appeal to Emotion and Wishful Thinking:
http://en.wikipedia.org/wiki/Appeal_to_emotion
By the way, your example of an implication in the TakeOnIt FAQ is not a logical implication.
How so? (one of those cases where something that should be obvious to me isn’t)
The second law of thermodynamics could be false.
Got it. I have changed the term everywhere from “necessarily” to “presumably”.