Feeling for Figures, by Mike Smith


At the risk of murdering, I’m going to dissect a piece of story.

My intention is to examine the way prose fiction works. My knife is the metaphor of the emotional button.

Could it be that stories work by pushing a series of emotional buttons in the reader? Words are the fingers, and mostly they push gently, raising no more than the consciousness of meaning. At others they prod more vigorously, arousing responses like trepidation and fear, anger or compassion; feelings of titillation or excitement. Every now and then they give us a multi-fingered stab that brings on those feelings of terror and pity that Aristotle talks about in his theory of poetics.

If that is the case, then we should be able to detect the process at work in any piece of writing, unless it is so incompetently written (or we are so obtuse) that it raises in us no understanding or emotion whatever!

I picked a book at random and flipped it open. So, my text is the first three paragraphs of Angela Locke’s novel, Dreams of the Blue Poppy (Hale,2007). This is an overtly emotional book, a three tissue weepie (even on a second reading!), so if the theory didn’t work here, it would be a trash-can exercise.

I considered the words of the three paragraphs sentence by sentence, after all, that’s how they would be read normally. I’ll take you through the first of those sentences in detail.


‘It was not until her grandson, Charles, had been born, on that April night, thirteen years ago, that Maud had felt hope was reborn again in that frail creature’ (Locke, p28)


No Hemingway short sentence here, at 30 words, but of them I make only 7 of heightened emotion (HE), and only one of very heightened emotion (VHE). That leaves 22 that are only tickling our responses, giving us meaning and understanding, setting up the HE & VHE words to be understood. These are the ‘its’, ‘that’, ‘the’ and so on, the logical bones of language. The other eight are worth looking at in some detail. By my reckoning they are: grandson, Charles, dramatic, night, hope, reborn, frail and creature. Their placing in the sentence makes them effectively into pairs of words, which I think enhances, increases, their emotional impact. ‘grandson’ and ‘Charles’ are bound together, rather than separated, by that comma, which, like a sleeping policeman, slows us down as we pass over them. Similarly ‘April’ binds ‘dramatic’ and ‘night’, adding its pennyweight to their values. ‘was’ fulfils a similar function for ‘hope’ and ‘reborn’, and of course ‘frail’, which was my only VHE word, stands directly in partnership with ‘creature’. In fact, in some ways those four pairings, taken in order, are a story in themselves, almost the story of the book, certainly the bones of its mainspring.

Their story, ‘Grandson Charles-dramatic night-hope reborn-frail creature’, gives us character, setting, and trajectory, the emotional threat in ‘frail creature’, coming after the high of ‘hope-reborn’.

The second sentence is shorter, at 14 words, and I found three HEs in it, but no VHEs. They do appear in the first six words of the sentence though, that intensity powering them up, and also throwing its weight hard against the emotional strength of that final HE-VHE combination which ends the previous sentence. Taking the two sentences as a pair, this gives us a swell of emotion within the larger paragraph.

That paragraph has two more sentences, a ‘quiet’ one of 14 words in which only one word poked its HE head above the surface, and a final one of 21 in which I found four VHEs. These were ‘trapped’, ‘crippled’, ‘prisoner’ and ‘fluttering.’ Again, I was struck by the story implied in these four words alone, and their button pushing was strengthened by the HE words that supported them: ‘Charles’, ‘ever’, ‘bath chair’ (treated as one), ‘ever’ and ‘heart’. In fact, almost half the words in this sentence are emotionally active, being HE or VHE. This is a highly charged sentence, and ends the paragraph on another emotional crescendo.

What I think this shows is that patterns of rising and falling emotional button pushing can be detected at both the sentence level, and at the paragraph level. This was an opening paragraph of a chapter too, and perhaps the place for a powerful start (or re-start, taken in a book-level context). In fact, when I went on to look at the next two paragraphs, both with more, and generally longer sentences, I found them quieter, with VHEs and HEs only sparsely scattered throughout.

By the time I’d got the resulting grid of statistics out though, I was beginning to doubt the value. Another reader might not identify the same words, nor score them as I had. Was I being consistent even, let alone objective in my scoring? I wanted to give words both higher, and lower scores, in subsequent paragraphs, than I had before. Short sentences seemed to make their words stronger. Clusters of words seemed to do the same. The effects of language are cumulative. Much humour, for example, works by incremental additions, not terribly funny in themselves, but riding the waves of laughter already created. Get an audience, or a reader, really laughing, and the most banal, even serious things, will seem, to them, to be hysterically funny. So the emotional content of the words cannot be viewed as stable and intrinsic, but must be, to some extent, a product of context.

Yet there is no doubt, spikes of emotion were identifiable in both sentences and in paragraphs, and even over the three paragraphs I looked at, the same rising and falling waves could be detected in the chapter as a whole, and they were created by the emotional loadings of individual words. I have no doubt that the chapters too, in the context of the whole book, would show the same patterns.

That was when I realised that though the statistics do not matter in themselves, the process of acquiring them had made me pay attention to the way in which the words create the emotional experience of reading the novel. More than that, the absence of consistency and objectivity which I felt had undermined the analysis was precisely where its value lay, for it was the effect of the pressing of my emotional buttons that I was in fact studying, rather than the activity of the presser!

The data produced are useless, but the lessons might not be. They are that it is a minority of words that do the job of story, and that the majority of words are in a supporting role; that combinations of words will have an impact greater than the sum of their parts; that the patterns of emotional intensity will be created at every level of the writing, of sentence, paragraph, and chapter. The numbers are not important, but an understanding of how we extract them, and why we perceive them to be there, is.