Consider the post Ads Don’t Work That Way by Kevin Simler on Melting Asphalt. It contrasts a couple of different mechanisms by which advertising might work, including:
Emotional inception—manipulating your feelings towards a product via conditioning
Raising awareness—making you aware of a product or its features
Cultural imprinting—creating a societal association for a product, to allow people to use the product to associate themselves with this same association
The details are described in the post, and I’m sure that Kevin Simler describes the theory much better than I could. However, they are not very important, as I am not bringing up this post to make a point about advertising. Instead, I want to zoom in on an argument Kevin makes:
UNDERHANDED ADVERTISING
Consider this one for Corona:
Whatever’s going on here, it’s not about awareness, persuasion, promises, or honest signaling. In fact this image is almost completely devoid of information in the most literal sense. As Steven Pinker defines it, information is “a correlation between two things that is produced by a lawful process (as opposed to coming about by sheer chance).” In this case, the image is so arbitrary that it can’t be conveying any information about Corona per se, as distinct from any other beer. Corona wasn’t specifically designed for the beach, nor does ‘beach-worthiness’ emerge from any distinguishing features of Corona. You could swap in a Budweiser or Heineken and no “information” would be lost.
This seems obviously true. After all, what inherent connection could there possibly be between Corona and a sunny beach? In a footnote in the post, Kevin admits that there is a slight cultural/circumstantial/contextual factor, but that’s not enough to change the core point of the post; the Corona/beach connection is obviously fairly arbitrary.
Except… now, years later, I’ve watched this YouTube video on the channel called “Food Theory”:
In it, MatPat explains that beers can get “skunked” when hit by sunlight. However, it turns out that Corona beer is uniquely designed such that skunking improves its taste, unlike other beers which it worsens. Therefore, there actually is a genuine non-arbitrary connection between the Corona brand and drinking beer on the beach.
This was a completely random fact that I hadn’t expected to jump out. I guess reality has a surprising amount of detail. But it makes the development and description of these broad social science theories annoying and difficult.
Obviously Kevin Simler didn’t know about Corona’s skunking properties when writing his post about advertising, and I didn’t know about them when reading it. If his readers had said “This example is unconvincing, because you haven’t shown there to be no special relationship between Corona and the beach.”, there’s not much he could have done; there could be any number of connections, and it’s hard to know if you’ve enumerated all the important ones. So it would be really hard to disprove the objection. But at the same time, as we can see now, the objection would have been right.
One solution to this would be to become an expert on everything you talk about, so you know even the most obscure potentially relevant facts. But that’s not practical. Another solution would be to accept being wrong—but I often get the impression that these sorts of things lead to people being wrong a lot. A third solution is to talk about narrower things, so you can specialize more into getting them right. Alternatively, maybe we could keep making sweeping generalizations, but only from things we have a lot of familiarity with? I don’t know what the best solution is, and it’s probably highly context-dependent, but it seems like a problem to consider.
Thanks to Justis Mills for proofreading and feedback.
I am quite happy about this post. It appears to be easily digestible by readers, and it points to a very important problem.
Rationalists can be great, but if there’s one thing we’re vulnerable to, then it is probably to get carried away with theories about things that turn out to be detached from reality.
One type of work that I would like to see more of in the rationalist community is comprehensive empirical work, to help work against the problem described here. Of course, it’s much easier to ask for that than to provide it. It might also be good to develop more rationalist theory about how to efficiently pick the relevant empirical things to work on.
Arguments are like boats. Very few ways of making shapes out of metal and wood will sail and won’t sink. Likewise, very few arguments have predictive power and survive scrutiny.
The difference is that it’s expensive and physically tiring to build boats, and there’s no benefit to building a useless boat, so most people don’t do it unless they plan to do it right. By contrast, it’s fun and easy and selfishly beneficial to craft and publish arguments, even if they’re sloppy. So people do it all the time.
You want to know how to build seaworthy arguments, and avoid climbing into leaky ones.This is difficult, because the overwhelming majority of the arguments you’re going to find docked in people’s blogs are of the leaky variety.
Unfortunately, checking the facts is often a thankless task. The author almost certainly won’t thank you, and the audience has usually moved on by the time you get to it. “A lie can travel halfway around the worldwhile the truth is still putting on its boots.” It’s fun to disprove arguments, though. On this level, I think it’s usually best to just assume in your gut that most things you’re reading are somewhere on the spectrum of wrong, mostly in inverse proportion to adjacency to math and the hard sciences, the author’s specific expertise, and the page count of the document and especially its bibliography.
Once you have a read on those factors, decide which ones it would be the most fun or useful to try and disprove. Occasionally, you’ll be pleasantly surprised. You might even find authors who consistently surprise you by being frequently accurate.
A second option is to patch up the argument, which I’ve attempted to do in another comment here. If the original argument is flawed, is there a piece of it that can be salvaged, reinterpreted, cut away?
A third option is to just focus on making your own arguments sturdy and true. This is difficult, for the same reasons that engineering any useful new product is difficult. You have competition, reality is surprisingly detailed, your customers don’t always know what they want, communication and cooperation is hard, and incentives are often misaligned.
[1] Corona. As Greg Rader and Caroline Zelonka have pointed out, the association between Corona and the beach isn’t wholly arbitrary. Corona is a Mexican beer, originally consumed (by American tourists) primarily in beach towns — so, as Greg puts it, there is at least a “circumstantial” basis for the association. Even for other products, the emotional or lifestyle associations probably have some anchor in reality.
I think the anti-skunking is more meaningful than this, because it’s genuinely a feature of the beer that makes it more suited for the beach, whereas the origin of Corona is more circumstantial. But YMMV.
I enjoyed the linked article, but thought its section on “why aren’t brands two-faced” was not argued correctly. It claims that the “inception model” predicts that brands would be two-faced.
From the linked article:
The inception model predicts that brands would benefit from being “two-faced” or “many-faced” — i.e., that brands ought to advertise to each audience separately, using whatever message is most likely to resonate with each particular audience, in order to provide maximum emotional impact....
One ad might link Gatorade to athletic performance, while another might link it to “having fun,” while yet another might play up its taste. Why not?
Not every aspect of branding can be altered for advertising to different consumer identities. Gatorade has control over which consumers see which advertisements, but it has to keep the name, bottle design, and label consistent. This is a weaker argument. There are plenty of lifestyle brands that have names and packaging that aren’t obviously adapted for only emotional/lifestyle niche. Corona beer is an example.
More importantly, emotions aren’t freely additive. We can’t add intense athleticism, relaxing beach vibes, intellectual fascination, coziness, and an anxious need for security and get an emotional mixture more effective at driving purchases than the sum of its parts. In combination, they have the opposite effect. They mutually undermine each other. If they could be added, they usually already have been—and given a name and treated as a separate category. For example, “Christmas Spirit” combines coziness and excitement and child-like wonder.
When these constraints are in force, the “inception model” doesn’t predict that brands would do multi-faced advertising. Instead, it predicts that advertisers would repackage the same basic product concept for different emotional valences.
So I don’t think we can use the absence of many-faced advertising as a point against the “inception model.”
In fact, you could make an argument that the “common knowledge model” predicts many-faced advertising. Because while emotions aren’t additive, lifestyles often are. What’s more appealing, being a businessman and nothing but? Or being a businessman who also is a rock climber who also is a responsible dad? If the lifestyle model is true, shouldn’t we see lots of brands positioning themselves as “the brand of every cool and virtuous thing you might possibly want to do?”
Nobody wants to mix the emotions of going to the Superbowl with the feeling of snuggling up to your partner in a winter cabin. But plenty people want to do both of these activities, at different times. And it’s common for people to develop a preference for a particular product, and bring it to both events. So why don’t we see brands trying to give businessmen the opportunity to subtly impress their rock-climbing skills and their responsible fatherhood upon their colleagues by the cheese plate they bring to the company picnic?
So it’s not clear whether the absence of many-faced advertising is a point against the inception or common-knowledge model, both, or neither.
My guess is that, as the OP here points out, advertisers are making implied arguments, which include common-knowledge but supercede it. It would be really dull to watch an advertisement explaining that Corona actually tastes better in the sun. But saying “Corona = beach” is a way of making this argument implicitly, and the consumer can test it for themselves.
The linked author really overstates the uniformity of colas. They don’t all taste the same. Even the fountain drink versions of the same brand of cola taste different from store to store. One feature of Coca-Cola and many other brands, like Doritos, is that they’ve managed to figure out a way to keep the product tasting uniform across the world and from year to year, despite changes in agricultural output that affects the ingredients. They pay people to ensure that Doritos get a mixture of corns that will keep them tasting the same each year, despite changes in the corn crop. With products that people eat frequently, those differences will be detectable. These companies wouldn’t put so much effort if “a cola is a cola.”
Common knowledge is another type of argument. “This wine tastes and is priced like the sort of thing it’s appropriate to bring to an upper-class family gathering” is the quiet part, and good ads help you convey it without saying it out loud. The common knowledge aspect adds a pleasant touch in which your contribution of wine is received by the host saying “I’ve heard of this wine, it’s supposed to be really good!” rather than “thank you for this wine, I’ve never heard of it before!” But pulling this off sustainably isn’t accomplished just by telling a consistent story that it’s “a good wine for upper-class family gatherings.” The advertising is promising that the wine really does taste good, and is expensive enough to be polite. That’s an argument, and the product has to be good evidence in support of it.
So my explanation is that ads are basically still making arguments. Emotions, common social knowledge, and many other factors can all be points in the argument.
Advertisers have just found ways to imply the argument more subtly, which makes it easier for a mass audience to take in. Emotional valence is part of that argument that helps argue for the product category (“Going to the beach? You’ll want to relax with a beer!”); common knowledge argues for the social reception (“Nobody will look at you funny if you bring Corona to the beach”); and the specific choice of how to select emotions and social contexts has to do with the product itself (“Corona is designed to taste better in the sun”). Accomplishing that in three words and a picture? Good advertising.
Another solution would be to accept being wrong—but I often get the impression that these sorts of things lead to people being wrong a lot.
Ideally, of course, you are always properly calibrated. In the real world, one solution might be thinking of it as your first and second order prediction, or Believing vs. understanding or a framing like in confidence all the way up.
I am quite happy about this post. It appears to be easily digestible by readers, and it points to a very important problem.
Rationalists can be great, but if there’s one thing we’re vulnerable to, then it is probably to get carried away with theories about things that turn out to be detached from reality.
One type of work that I would like to see more of in the rationalist community is comprehensive empirical work, to help work against the problem described here. Of course, it’s much easier to ask for that than to provide it. It might also be good to develop more rationalist theory about how to efficiently pick the relevant empirical things to work on.