A Poetic approach to collecting human-centered data

Call me Ishmael. — Moby Dick, Melville

This is a sentence that is prose. It is an objective fact. There is a person and their name is Ishmael.

But is it?

Why say it that way? Why not say “My name is Ishmael?” Is this person hiding something? Was Melville just creating a sense of mystique using vaguely biblical language? Is it just a way to start the book?

This sentence is part of a larger piece that is not just a collection of objective facts. It’s not just prose. It tells us a deeply moving tale that makes us feel and understand more than just a list facts.

Language can exist in two places at once:

A prosaic level — these are the facts.

A poetic level — this is a sensory area where the words and the structure cannot be separated from aesthetics or emotions.

I think many people can intuitively relate information using some combination of the two. Perhaps one person is a little more “matter of fact” and talks openly about the minute details of what happened to them and how they felt about it. Someone else turns to using anecdotes and metaphors to make a deeper point that they hope is self evident.

I believe many interviewers, with very little training, can remain easily on the prosaic level of an interview. It is easy to consider: what facts are they telling me? What facts do I not know yet but would like to? Is there more to a fact?

“Does Jack know Jill?”

“Yes.”

Many interviewers can intuit the poetic structure. They understand that emotions have taken place or are being felt now; people are somehow relating to the story being told. Often a journalist can sense if a person is being deceitful, or spinning facts, or are being aggressive. Perhaps the journalist themselves are trying to pin the person down and get them to admit something, like in an interrogation. Intuit or not, I have seen few interviewers know what to do with the poetic structure, not to even mention having any control over it.

Asking about emotions brings us closer to the poetic level however I would argue that asking someone how they feel is still fact based. It stays on the prosaic level.

“How does Jack feel about Jill?”

“Does Jack love Jill?”

This is approaching more complex territory but it is a prosaic approach.

A poetic approach to an interview would be to consider the emotional and aesthetic presentation of the interviewee and how the specific words and structures of the sentences contribute. Do they sound like a reliable narrator? If they are an unreliable narrator, how does that change the questions to be asked?

The poetic level — this is a structure that sits above the prose to give it a meaning that is deeper than what the prose suggests. It is defined by how a person transforms a base sentence.

Here is a base sentence. No frills:

I am tired.

Shakespeare transforms this base sentence into:

Tired with all these, for restful death I cry. — Sonnet 66

A millennial might transform the base sentence into:

I can’t even.

In both of these transformations the idea stays the same but the words and the structure change and, in so doing, you can tell something about the people who say these sentences. There is a new meaning that is in addition to the fact that the speaker is tired. This new meaning sits on the poetic level.

What’s necessary is a careful, ongoing, analysis of what is said on a prosaic and poetic level. A person might say something that is meaningless on a prose level; complete gibberish. But on a poetic level, this gibberish not only makes sense, but is revealing of a deeper truth. A business owner once told me that she expects her employees to be: quick thinking, observant, able to improvise, but not all the time. “It’s situational.”

I asked, “what are the situations? When are they expected to be this way and when aren’t they?” She said some more words, something about multiple locations, before allowing her sentence to slowly drift away.

On a prose level, there was nothing to engage with. I could have forced a question: asking her to clarify, but it wouldn’t have accomplished anything. On a poetic level, she was displaying confusion and answered in a way that said she did not relate to the question at all.

What’s revealed from a poetic level is that she can assertively answer that she does have expectations from her employees. However, she is less assertive with it being situational. She misspoke and I pressed her on the wrong point. I asked the wrong question. I asked a prose based question: “what are the situations?” I only realized this was the wrong path from looking at what she said on a poetic level.

People do not only speak in either prose or poetry. You shouldn’t wait to hear gibberish to consider the situation from a poetic level. People will explain facts, tell tales, share anecdotes; that is all prose. They poetry is the exact words they say and the structure they use which tells you how they are relating to the story.

A person could tell a story in a long, circuitous way; going on tangents, dropping the story in one question only to return to it later on a different question. They take pauses. Ambling sentences. They story isn’t refined because they might be telling it for the first time and are finding the words for it. They might be writing it, so to speak, in their minds and telling it to you live. They are wrestling with it, and the events you are asking about have not yet been fully processed yet. It might be that if you return to this person the following day or the following week, they could have reflected on it more, internalized it differently, and tell you something completely new.

Counter that with a person who has a very smooth story. They are confident in how to tell it. They have some jokes. The story slides off their tongue. A poetic analysis could reveal that this person has told this story before, possibly countless times. This story to them now, at this moment, doesn’t have any emotional charge except that they know that they can engage and entertain a listener with it. The insights are rehearsed. It could be hard for them to consider this story from a different angle. It might be hard for them to drudge up something new in the well trodden path of their story. Politicians are sometimes accused of “dodging questions”: when they deliberately answer a question with a planned response that is meant to exude confidence but also downplays and hides damaging details. That is a poetic criticism, and the fact that they didn’t share the appropriate details is a prosaic criticism.

So does this mean that we only focus on the poetry? Is it that the meaning, the true meaning, comes from listening to how a person says something not what they say?

Of course not.

Listening just to the poetry would be similar to listening to someone speak in tongues, or proclaim something very emphatically but in another language. It would tell you something but at this extreme would also be meaningless. Deriving anything would be an exercise of the interviewers creativity; turning nonsense, near-fictions, into claims of fact. Basic conjecture.

The prose gives structure to the poetry. The facts and events that the person is recounting act as a basis which then the person uses to put emphasis, stress details, or act in a way that enhances the facts. This emphasis, stress, and act should also give you an idea of not only what to ask next, but how to do it.

For example what if the interviewee is bored? In fact they are bored by you, the interviewer. Could you tell? Should you keep asking questions and march dutifully onward through the interview guide? At best you’ll get honest yet utterly disconnected answers; from which you will later have to derive “user needs.” Doubtful. At worse, you’ll get dishonest answers or none at all.

Boredom is an easy one tell. What if the person is being sarcastic? Or aggressive but still personable? Like they are joking and having fun but the jokes have a little bit too much of an edge to them. That one is more challenging.

However there is something even more challenging. What if the interviewee is feeling pressured to answer? They have no relation to the question, so they come up with an answer. They are being nice; trying to give some answer, any answer, to the interviewer as a gift.

“Oh my, Jack and Jill, but of course! Let me tell you I’ve thought of these two since I was a child. To be honest, I’ve always thought of myself as a Jack.”

That isn’t serious.

This ability to look at the prose and poetry of an interview is just that: an ability. It is not necessarily the data. This ability helps interviewers elicit a more complete picture of the truth and perspective that the interviewee holds, even if the interviewee doesn’t necessarily tell it outright.

I think there is a way to consider the data you are collecting that lends itself well to doing this ability — looking at the prose and poetry — more naturally.

One way to look at human-centered data is to believe that doing the data collecting, be it interviews, questionnaires, whatever, is a way to assemble a set of facts. That is definitely what a prosaic approach to interviewing will provide.

Jack said this, Jill said that, and Gertrude said something totally different.

Ideally this set of facts lets you create some causal or correlative chain, like a cause-and-effect chain of events. Overall, this approach is similar to what an investigative journalist might do. This works, but relies on you having access to the right people at the right time to interview. Often it isn’t even clear if you are interviewing the “right person” or you might not even know who the “right person” is, which is basically what Deedee Gordon says.

I think a more helpful way to think of data, is that when you are collecting human centered data, you are building a knowledgescape — a word coined by Perry Zurn and Dani Bassett. I am going to avoid using words like systems and networks (although you can think them if you like; these words do apply here). I’m avoiding these words because it’ll get us trapped in technical jargon and ideas like levers, patterns, hysteresis, scale-free, complex vs. complicated, and I will defer to Jeremy Lent and Donella Meadows for that discussion.

No, knowledgescape is a very appropriate word because it immediately conveys the right idea: there is this universe of knowledge and people and concrete objects and abstract values all make up this field that you (the researcher) get to revel in. In the knowledgescape surrounding our running example we know that Jack is important and so is Jill. We know the concept of love is also a component in this knowledgescape. Gertrude is floating around somewhere. Some things are big and important. Some things are small and insignificant, yet still there. These are all components in this knowledgescape and in the course of our data collection we get to add in components as we hear more. We also get to understand how these components relate, if they even do.

How does Jack relate to love? How does Jill? Is there a relation between Jack and Jill other than through this mismatched love component? How the hell does Gertrude fit in? Some components are close together like Jack and love and Jill. Some components like Gertrude are isolated.

In thinking about data collection this way, it becomes more natural to turn to balance the prosaic and poetic approach. The prose gives you the components. The poetry in an interview allows the researcher to glean how an interviewee might relate to another component floating out in the knowledgescape.

I think this is closer to what Adam Smith was writing about:

Curious attention labors to connect one thing to another. — History of Astronomy

I’ll close with this thought.

“Call me Ishmael” is not the first line of Moby Dick. The entire first section of Moby Dick is an encyclopedic journey into whales conducted by the Pale Usher and a sub-sub-librarian. Together, they go into whales as depicted in art, the etymology of whales, the phrenology of whales, the taxonomy of whales, etc.

The point is that over the years the whale has taken on a lot of meanings. How the whale connects to these components, and what those components have to do with the deranged Captain Ahab, the narrator Ishmael, and the doomed ship The Pequod, is up to you to figure out.

Prose and Poetry.