Chapter 86: Consider the Nut Graph

With apologies to Jane Austen: It is a truth universally acknowledged that a true story in possession of ambition must be in want of a nut graph.

The nut graph, for those who have never had to write, rewrite and rewrite one again, is the paragraph whose job is to anchor a story, explaining to readers what the story is about, why it matters, the context in which it occurs, and why, as a result, they will want to read it. Nut graphs serve a useful purpose in that without them stories can feel like they are floating. Nut graphs also force writers to do what so many struggle to do: commit. To say, with clarity and certainty, This is a story about…

So sacrosanct is the nut graph that in the years when The New York Times was primarily a print publication there was an iron clad rule that stories that began on Page One had to include the nut graph before the story jumped inside. Google the term nut graphs and the page will open to an array of suggestions on how to write good ones, why they matter, what they accomplish as well as the inevitable iteration of Five Tips for Success. 

I do not recall the term nut graph being invoked when I was starting out in the mid-1970s. But the spirit was there, couched in less catchy phrases like “the billboard” or “the big picture.” Count on journalists to come up with a shorthand that both conveys the essence of an idea (here is the nut of the thing) and, in the spirit of affected grouchiness with which so many of us go about our work, also offering a sly, double meaning. Because, nut graphs can, forgive me, drive you nuts.

It is not the nut graph, per se. Rather it is the way some editors approach it. Nut graphs can take up a disproportionate amount of an editor’s time as they refine, tweak, re-work, re-cast, punch-up, expand, tighten, and oftentimes beat the living daylights out of a nut graph until it feels right. Editors, like nature, abhor a vacuum, and there is little that sends them racing for a red pencil (or Track Changes) faster than a flabby, imprecise, meandering nut graph. 

What are we saying here? they will ask.

To which a writer had best be prepared with an answer that does not include such words as – kind of, sort of, seems, or appears. 

Well, why didn’t you just say that? an editor will reply to a writer whose spot-on nut graph was there all along, waiting in their head, but missing from the page. 

I am an impatient reader of the news; I do not want to spend a lot of time with throat clearing before I know what’s going on. So when it comes to conveying information, I am all for the nut graph and grateful to every editor who labored to make one better. 

The problem with nut graphs is not  their presence in keeping readers caught up on events. Rather, it's when that same impulse to explain fully and completely is applied to storytelling. Because then that otherwise vital, often essential paragraph is transformed into something quite different: buzz kill.

A couple of weeks ago I conducted a very unscientific two-part experiment with my students. For Part One I shared with them the opening 500 words of a story that had recently run in an Important Magazine. The story involved a crime. There was a lede and nut graph. I asked how many would have read on. Two raised their hands. The other thirteen passed. 

Why? I asked, rhetorically. In truth, I had shared the opening with them to see if their reaction was similar to mine – that just as the tale began to unfold, the writer (in cahoots no doubt with an editor) had decided to stop the narrative in its tracks and provide context. Context? Now? Just as we’re getting rolling? Are you kidding me?

This sin of commission was not an outlier. A generation of writers and editors have embraced the idea that a story that begins anecdotally must pause to include a nut graph. And because this often occurs in stories of some length, the reasoning goes that the nut graph, too, should be equally expansive -- not the quick, trenchant hit of a news story, but an extended paragraph that all too often includes such phrases as – What happened to John when he committed the most daring bank robbery in the history of thievery is part of a trend that sociologists term….

Part Two of the experiment was something I’d done before, a game called “Fiction or NonFiction?” I read the beginning of a story, stopped and asked the class which it was. The answers were not always obvious. There were true stories they were sure were fictional. There were others they were positive were nonfiction. The difference: the appearance early on of a big, idea/fact/context laden nut graph. For the record, they were shocked to discover that Gay Talese’s legendary profile of Joe DiMaggio was not a short story. It was a story, wonderfully, blissfully uninterrupted by the journalistic equivalent of Eat Your Broccoli. 

Like so much in journalism that feels it has been part of our tradition since Moses took dictation from God for the story that would become The Ten Commandments, the sanctity of the nut graph was not always so. If you are a writer of a certain age, you took inspiration from the 1973 “The New Journalism” anthology. And the experience of leafing through my now well-worn copy is one of rediscovering story after story with nary a nut graph in sight. So what happened?  If those Hall of Famers didn’t include nut graphs, why do we cling to them? 

My theory: the contemporaneous rise of the anecdotal lede. 

The anecdotal lede – and I say this as someone who has written maybe a few thousand – is in effect a statement, a pronouncement to readers that the piece they are about to read will unspool not as an article but as work of storytelling. Sadly, however, too often those ledes are a mere approximation of the experience.

A show of hands for everyone who has written an anecdotal lede that was fine (mine is raised) but somehow failed to seize readers by the collar and say, you need to read this. Yes, the story opened with an anecdote. But – be honest – was that anecdote tentative or vague? Was it unproductive – meaning did it leave readers unsure of what the story was about, where they were being taken, and whether their investment of time was worth it?

When a story feels as if it has to open with an anecdote – which has evolved over the years into the default lede for stories that are not pegged to the news – too often the anecdote can end up feeling forced. There is something more: a tentatively rendered anecdote is often a way for a writer to avoid committing to a direction or idea. The words serve as camouflage for a writer unsure of what she or he wants to say.

So what became the remedy for such indecision and meandering wordiness? The nut graph. But in addition to bringing what drama there was to a grinding halt, the nut graph effectively undermined the lede by saying, okay we’ve set a scene and told the beginning of what feels like a story but we’re guessing that you are feeling a little lost now so let me explain to you what’s to come.

The effect is akin to a joke you need to explain. If you need to explain it, it’s because no one got it.

So are we doomed to live with the narrative-killing nut graph? I don’t believe we are. But how then to open a story with an anecdote – a scene, a character in motion, a moment – that does not leave an editor thinking, we really need to punch up the nut graph? 

The answer, I believe, can be found in the guiding principle of “search.” Google did not invent a new form of human behavior with search. Rather it took a neurological function that has always existed and applied it to a vast storehouse of knowledge. To access that knowledge, you enter into the search field certain words whose job it is to dive in, look around and return a nanosecond or two later with results. In other words, the search terms acted as triggers, prompts.

It is the same with storytelling. By adding a few choice but essential words to the first, say, twenty five words of a story, a writer hopes to spark associations for readers. And those associations, in turn, will produce a cascade of further associations through which the reader will quickly see and, yes, understand the context for that story.

Consider two fictitious anecdotal ledes: 

Exhibit 1: Sally Jones, 53, always wanted to be an opera singer. Her parents had encouraged her and she took voice lessons all through her childhood. When it came time to audition for music school, however, Sally was not accepted.

“I was so disappointed,” Sally said. “I guess I was just too nervous.”

Sally’s experience echoes those who suffer from stage fright. In a 2012 study by…

Exhibit 2: Sally Jones felt herself sweating and knew she was in trouble. She looked out at the darkened theater and waited to be asked to sing. The wait felt interminable and now the sweat was running down her back and beading on her upper lip.

“Are you ready, Miss Jones?” called a voice.

Sally took a breath and began to sing an aria she thought she had mastered.

You take my point. The first feels rushed which saps the opening of drama or impending doom. It also lacks the keywords of the second: Sweat. Trouble. Knew. Wait. Interminable. In the second the context is built into the story. We get it. And better still, it connects us, the readers, to the story. We are in there together with the writer. And we are in for the ride: We want to know what happened to Sally. Suddenly we find ourselves caring about someone we have never met.

I want to close by inviting you to take part in five-second survey on nut grafs. It asks one question:

We’ll let you know the results next week?

Thanks