The art of persuasion 2: Barthes and the audience (for marketing)

A Barthes sign – deliberate irony… So, last time I started at the beginning of the art and science of persuasion, looking at how Aristotle’s rhetoric is still the basis of modern marketing, via his 3-part system of:

  • Ethos (basically the speaker’s credibility or brand)
  • Logos (the subject & style of the message)
  • Pathos (effectively the audience’s response)

Prioritising pathos

For an artist, novelist, or even journalist, the first two of these are almost always the most important, with a strong emphasis on the second. While most artists and writers *want* to have a positive audience response to their work, their primary goal is to get their idea out there. Their ethos / reputation may help them reach an audience, but this is secondary to the logos of the work they’re producing: the subject they want to convey, and the way they – the authors of the work – decide to shape it is both their focus, and the focus of their audience.

In short, for an artist or writer, often it is the act of creation itself that is the goal. Getting a positive audience response is merely a bonus – and being forced to chase an audience can stifle their creativity and lead to both stress and bad work. Hence the cliche of the difficult second album…

By contrast, for marketing it’s the last of Aristotle’s three concepts – pathos, the response of the audience – which is most important. Without the right kind of audience response, a marketer’s work will have failed. Hence the birth of focus groups, testing, and the often quite derivative nature of advertising, as “creatives” are forced to shape their work not around a great creative concept in itself, as an artist would, but how that creative concept is likely to resonate with their target audience – pushing them into creativity by committee (always a killer), and a constant recycling of ideas that are known to work.

This focus on the audience’s response is how we’ve end up jumping all the way from Aristotle to Roland Barthes, the notoriously difficult to read late-20th century French semiotician. (But still, at least he’s not Pierre Bourdieu, who I may get to in a later piece in this series…)

Everything is subjective

Barthes’ most famous idea is that of the death of the author – basically the idea that the authorial/creative intentions behind a work of art/literature don’t matter; all that is important is the response of the audience.

To understand this response, Barthes – building on some of the concepts of Ferdinand de Saussure that helped give birth to semiotics, the study of signs (combined with some of Jacques Lacan’s ideas on psychology) – began to re-conceptualise the way meaning is created in culture and society.

Where for Saussure, the importance of semiotics was to help understand the connection between a symbol/sign (the signifier) – be that a sound, a word, or an image – and the thing or concept that symbol/sign was intended to represent (the signified), Barthes effectively took this one logical step further by pointing out that this two-way connection still didn’t get to the heart of the *meaning* of that sign/symbol, because meaning is entirely a matter of interpretation. The connection between signifier and signified is entirely subjective.

Why? Well, because all of us have different knowledge, experience, ideas, attitudes, needs and expectations.

Emojis and meaning

Emojis are a good example. Take this one: Culturally confusing dumpling emoji Originally designed to represent a Chinese dumpling, the emoji’s creator specifically had dumplings in mind when she made it, arguing:

“The dumpling is actually universal. Georgia has khinkali. Japan has gyoza. Korea has mandoo. Italy has ravioli. Polish people have pierogi. Russian people have pelmeni. Argentians have empanadas. Jewish people have kreplachs. Chinese people have potstickers and various other dumplings. Tibet and Nepal have momos. Turkish people have manti.”

As emojis are intended as shorthand signs to speed up communication, arguing for a dumpling emoji based on that logic is pretty sound.

When I first saw it, I initially thought it was a Cornish pasty – a type of food I grew up on. This wasn’t explicitly included in the creator’s initial list – but it was there in the spirit.

In other words, a symbol that means one thing to one person could mean something very different to someone else. (A bit like when my mother kept on signing off text messages with “LOL”, meaning “Lots OLove”, and I was reading it as “Laugh Out Loud” and wondering what was so funny.)

Back to Barthes and a multitude of meanings

Anyway, all this is to illustrate one of Barthes’ key points: Meaning isn’t as simple as there being a direct connection between a signifier (word/image/sign) and signified (thing/concept). Instead you also need to consider the interpretation of both.

This led to Barthes’ modification of Suassure’s bilateral signifier-signified relationship into a trilateral model:

  • Representamen: the signifier / sign / word / image / sound used to represent a concept or thing
  • Object: the signified / thing / concept being represented
  • Interpretant: the person decoding the meaning denoted by both Representamen and Object – a meaning that may vary wildly from interpretant to interpretant depending on their personal context

In other words, there is no single objective, definitive, “correct” interpretation of any given representamen, because the meaning of such signifiers (and even the objects/signifieds they are intended to represent) will constantly change according to context.

An alcoholic example

Take the phrase “I want some alcohol” to illustrate the point.

In one sense, “alcohol” always means the same thing – reading that word, you *think* you know what I mean by it in writing it down. But – as with the emoji example above – what kind of alcohol is, to you, representative? Beer? Whisky? Wine? Gin? A fancy cocktail with an umbrella in it?

A selection of different types of alcoholic drinkIn Japanese, the word for alcohol is sake – which is fairly familiar to English speakers as Japanese rice wine. Ask for sake in a Japanese bar (at least, if you’re obviously not Japanese, like me), and that’s what’ll be delivered, even though the word technically means all types of alcohol.

But what if you’re a surgeon, talking about medical alcohol to clean your scalpel? Or Muslim, and alcohol is forbidden by your religion? Or an alcoholic, and the very thought of it represents a constant temptation and potential relapse? Or, to a lesser degree, what if it’s nine o’clock on a Saturday morning and you’ve got a hangover?

And then there’s the time and place (kairos, for Aristotle). “I want some alcohol” if you’re the surgeon in the operating theatre above will be purposeful, urgent, obviously related to a specific medical need. Say it at the end of a long working day, it could be an sign that you’ve worked hard and deserve a reward, or that you’ve been worked *too* hard and are feeling depressed, or just that you’d like to spend some time with your colleagues in a social environment. Say it at seven o’clock in the morning, and it likely means you’re an alcoholic (unless you’ve been going all night, in which case it means you’re a bit of a party animal – unless you’ve been going all night and you’re in your forties, in which case it means you’re having a mid-life crisis, and probably a little bit sad).

Even if both you as author and interpretant as reader are agreed roughly on what specific type of alcohol you’re referring to, there are still additional contextual meanings that the interpretant will layer on top of your representamen, often subconsciously, that will trigger very different responses.

Again, Aristotle got in there first – because this is pretty much what he had in mind with the idea of pathos: the emotional response of the audience to the rhetorical approach you, as speaker, have taken.

It’s always about your audiences – plural

So, if everything is subjective, context is everything to interpretation, and the intended meaning and interpreted meaning can vary wildly from person to person and context to context, what does this mean for effective communication?

Well, it basically means that it’s very, very difficult to communicate effectively.

But there is hope – Barthes doesn’t go as far as his fellow postmodernist Jacques Derrida (of whom more another time, probably) and argue that all this means things are changing so much there is ultimately no such thing as meaning.

But what it does mean is that we shouldn’t make assumptions about how what we’re trying to say will be interpreted. We need to think more about principles of inclusive design when starting to craft our messages. We need to constantly self-criticise and consider the vast range of cultural, linguistic, social, educational, and personal experiences of our audiences. And to recognise that audiences are always plural – even if the audience is a single person, because their mindset and mood will vary depending on the context in which they encounter the thing it is they are interpreting.

To anticipate all these variants may well be impossible. But if your job is to convey meaning and to try to persuade – as it is for marketers and advertisers – then your job is to attempt to anticipate as many as possible.

Anticipating responses – and adapting accordingly

Which is why, to be a successful marketer, a basic understanding of the media, of creative techniques, of the product you’re marketing and the sector you’re in – the traditional needs of the industry – is not enough. To be effective needs a far broader understanding of the context in which your work will be interpreted by your target audiences. And these audiences are far more complex than can ever hope to be summed up on a PowerPoint slide as a “persona”. To try and anticipate their responses needs an awareness of a huge range of potential variables – far more than the simplistic, old-school advertising approach focusing on fairly simplistic demographics.

This is why, as I continue this series exploring the art of persuasion and the theories and best practices that underlie marketing, I’m going to continue to branch out into other areas – from semiotics to sociology to anthropology to literary criticism to linguistics to economics to psychology to philosophy to history and more. Because ultimately, the only way to get close to persuading is to try to understand and anticipate the vast range of ways what we, as creators, are doing can be interpreted, and construct our attempts at persuasion backwards from this knowledge.

Barthes never quite got to an all-encompassing theory of meaning to create a roadmap of how to do this, having died mid-flow at the age of 64 after being run over by a laundry van following a convivial lunch with future French President Francois Mitterand and the philosopher Michel Foucault (of whom, probably, more another time). I doubt I’ll be the one to pick up the baton. And it may well be impossible anyway – even tapping into the potential promise of AI to analyse vast datasets and interpret correlations and likely causations of meaning and influence.

But hell – it’s a way to pass the time.

The art of persuasion series:

  1. Aristotle’s rhetoric: the foundations of modern marketing
  2. Barthes and anticipating audience responses


Black Lives Matter logoWords are important, because language shapes our understanding of the world.

Over time, our choice of language can shatter or reinforce preconceptions – creating feedback loops of frustration or moments of radical shifts in perception that in turn can change society itself, for good or ill.

The same is true about our choice of what to talk about – or to ignore. Sometimes, staying silent is as strong a statement as speaking out. Sometimes, speaking out is a risk.

But for those of us – people or organizations – in a position of privileged security or power, sometimes speaking out is a duty.

The question is, what message will you send about what you see as important in the world in the words you use and the things you choose to talk about? And what good could your words do when you do speak out?

  • This piece from NiemanLab shows that the choice of language in covering protests about racial inequality is yet another area in which society is unfair and promotes systemic inequities.
  • This tweet highlights the importance of action as well as words in supporting movements for equality – especially from brands.
  • This TED Talk is an eye-opening, amusing analysis of how the language we use and the way we frame discussions about racial violence can point to the absurdity and insanity of racist norms.
  • This call from my current employers for brands to take action as well as show their support was rather good – and there has been action at my place behind the scenes, not shouted about, that has made me rather proud of my colleagues.

When writing is a matter of life and death

Behind the Economist paywall, sadly, but this is the key point – always worth remembering beyond the current crisis, and something a number of political leaders need to learn:

“Recommendations that sound more advisory than mandatory seem to presume rational adults will do the right thing with accurate information. The central insight of behavioural economics is that they do not…”

Clarity of messaging is more vital now than ever – it’s become a literal matter of life and death. “Practice social distancing” is vague, confusing, advisory. “Stay home” is clear, unambiguous, mandatory. Guess which has worked better?

This is also a lesson we should take with us after this crisis has passed: If you want to be understood, make your point as clearly and plainly as possible. Otherwise you have no one to blame but yourself when people don’t pay attention to what you’re telling them – or, worse, start believing simpler-sounding misinformation.

The endless battle against “garbage language”

Complaining about nonsense business-speak may be futile, but this piece – a review of a memoir about life in startup land – does a good job of summing up why spewing out business bullshit is not just intellectually offensive, but actively harmful:

“I like Anna Wiener’s term for this kind of talk: garbage language. It’s more descriptive than corporate speak or buzzwords or jargon. Corporatespeak is dated; buzzword is autological, since it is arguably an example of what it describes; and jargon conflates stupid usages with specialist languages that are actually purposeful, like those of law or science or medicine. Wiener’s garbage language works because garbage is what we produce mindlessly in the course of our days and because it smells horrible and looks ugly…

“But unlike garbage, which we contain in wastebaskets and landfills, the hideous nature of these words — their facility to warp and impede communication — is also their purpose. Garbage language permeates the ways we think of our jobs and shapes our identities as workers. It is obvious that the point is concealment; it is less obvious what so many of us are trying to hide.”

In short, if your ideas are good, don’t bury them in garbage. If they’re not, the presence of garbage is a good indicator.

Web writing, hate reading, and the decline of quality

Nothing new, but this is worth a read on web writing and hate-reading – that old trick of being as controversial as possible in order to get an extreme response, purely because extremes get more attention, and in a pageview-driven business model, controversy is seen as good purely because, based on the metrics, it’s the controversial stuff that’s driving engagement.

This infantile attitude of provocation to get attention is increasingly being combined with ream upon ream of cheap content, because the more content you’ve got, the more potential PVs you can attract. We end up with the most depressing (and false) equation of online publishing:

Cheap content + Controversy = Clicks = Cash

It’s an attitude that’s lazy *and* massively short-termist in thinking – over the long term, quality can and should trump quantity. But even if it doesn’t, cheap, crappy content is a turn-off for audiences. The more sites that start to rely on hastily-produced, poorly-checked copy, or lazy semi-plagiarisms of things that desperate teams of poorly-paid hacks with deadlines and quotas to hit have found elsewhere, the less distinctive sites get, and the fewer returning visitors you’ll get. As that linked article puts it:

“With a business model based on a ton of cheap content, Web publishers can rely too heavily on acid-reflux-style aggregation, in which young writers destroy the savor of interesting stories and an interesting world by constantly regurgitating the news with added bile.”

There’s also an interesting point made from John Waters in the Irish Times (now behind a paywall), on the impact of comment sections under online articles: “Because everything written specifically for online consumption is written in the expectation of addressing a hostile community, the writing process demands, as a prerequisite, either a defensive or antagonistic demeanor.”

Having learned my online publishing trade in the realm of message boards, chatrooms and blogs, I’m incredibly aware of the vast levels of bile that exist in comment sections. But it doesn’t have to be this way. With careful community management, it’s perfectly possible to build online communities that are supportive, friendly, and constructive, rather than the supposed default of objectionable and offensive. Check out the likes of b3ta, imgur and Metafilter for some prime examples of sites with vast *positive* communities of commenters. And then contrast those with the comments sections of pretty much any national newspaper site – packed with trolls and maniacs.

It doesn’t have to be this way.

Odd numbered lists are good, says science

We still don’t know why, though…

“There are many more listicles of length 10 published compared to other numbers. This is primarily because BuzzFeed is selling the 10-length listicle to partner brands, such as the Michael J. Fox Show, Nordstrom Topman, and Buick. The second most popular length is 15, followed by 12. Listicle length drops off quite rapidly in the 20’s, although surprisingly, lengths 11-21 are far more popular than those under 10…

“If we look the bar chart by audience score we see a completely different picture?—?odd number length listicles… tend to have a higher audience score on average, where in our dataset, the number 29 tends to have an advantage over the rest.”

How to write clearly

I usually hate tips for writers – writing, to me, should be a natural thing. But having seen a lot of very bad writing, more concerned with showing off the writer’s linguistic skill or subject-matter expertise than enlightening the reader, this approach strikes me as vital to keep in mind at all times:

Writing is a modern twist on an ancient, species-wide behaviour: drawing someone else’s attention to something visible. Imagine stopping during a hike to point out a distant church to your hiking companion: look, over there, in the gap between those trees – that patch of yellow stone? Now can you see the spire? “When you write,” Pinker says, “you should pretend that you, the writer, see something in the world that’s interesting, and that you’re directing the attention of your reader to that thing.”

Perhaps this seems stupidly obvious. How else could anyone write? Yet much bad writing happens when people abandon this approach. Academics can be more concerned with showcasing their knowledge; bureaucrats can be more concerned with covering their backsides; journalists can be more concerned with breaking the news first, or making their readers angry. All interfere with “joint attention”, making writing less transparent.

This isn’t a “rule for writers”; it’s a perspective shift. It’s also an answer to an old question: should you write for yourself or for an audience? The answer is “for an audience”. But not to impress them. The idea is to help them discern something you know they’d be able to see, if only they were looking in the right place.