The most requested feature for anything that holds emotional material is the one we have decided not to build. People want it to tell them what it means. You have written about your mother nine times in six weeks — what does that say about me?
It is a completely reasonable thing to want, and refusing it needs a better answer than modesty. Here is the answer.
The interpretation is the valuable part, which is why it should be yours
Start with what an interpretation actually is. It is a claim about causation and significance in a life the interpreter has no access to except through a text box. You keep writing about your mother because you have not resolved something. Maybe. Or your mother is ill. Or she has been in town. Or you are writing a piece about your childhood. Or nothing — some months a person is just on your mind.
A system that says the first thing sounds insightful and has, in fact, guessed. The trouble is that the guess does not land as a guess. It lands as a finding, because it arrives in the voice of something that has read everything you wrote and can count.
And here is the part that makes it worse rather than merely inaccurate: you will probably accept it. Richard Nisbett and Timothy Wilson's classic review showed that people confabulate explanations for their own behaviour with great fluency and no awareness of doing so, and readily adopt plausible-sounding accounts offered from outside.1 An interpretation handed to you at the right moment does not compete with your own reading. It replaces it, quietly, and you will remember it afterwards as something you worked out yourself.
That is the harm. Not being wrong — being adopted.
What noticing is, and where it stops
There is a real difference between two sentences that look similar.
Your mother came up nine times in six weeks. Before that, twice in three months.
You seem to be working through unresolved feelings about your mother.
The first is a count and a comparison. It is checkable — you can go and look. If it is wrong, it is wrong in a way you can catch. It adds something you genuinely could not do yourself, because you cannot count across six weeks of your own writing from memory.
The second is a story about your interior. It is not checkable. It is not falsifiable by looking. And it does the one thing that was actually yours to do.
We will do the first and stop. The stopping is not a limitation we hope to remove in a later version. It is the design.
Naming and explaining are different operations
It would be easy to read this as a claim that nothing helps and you are on your own. That is not the claim.
There is good evidence that putting feelings into words does something measurable. Matthew Lieberman and colleagues found that labelling an emotion — simply naming it — reduced amygdala response, without any interpretive work at all.2 James Pennebaker's long programme on expressive writing found health and wellbeing effects from writing about difficult experiences, and the effects were associated with the writer's own construction of meaning over sessions, not with anything supplied from outside.3
Read those together and the shape of the useful thing appears. Naming helps. Writing helps. Arriving at your own account, over time, helps. Being handed someone else's account is not in the evidence at all — and Pennebaker's finding that the benefit tracks the writer's own meaning-making suggests that handing it over may remove the mechanism rather than accelerate it.
So the division of labour is not arbitrary. The system does the part you cannot do — holding six weeks accurately, in order, without your mood today colouring what it kept. You do the part it cannot do, which is knowing what any of it means.
The confidence problem
There is a practical reason as well as a principled one.
Something that has watched you for two years should be less certain about you, not more. Long records do not converge on a clean reading of a person; they accumulate exceptions. The months that contradict the story. The person who was the problem and then was not. The mood that had no cause you can find.
Any system built to produce interpretations has to suppress that. It cannot say here are nine things about your mother and they do not add up to anything I can see, because that is not a satisfying output, and a product optimised for satisfying outputs will find a story. The pressure is always toward more confident readings of thinner evidence.
We would rather show you nine things that do not add up. Sometimes they do not. That is a true fact about your life and it is worth more than a tidy sentence.
What this feels like in practice
Once a week you get a letter. It is short. It says what recurred, who was in the writing, what changed since last time. Where something shifted, it says what it was before.
It does not say what to do. It does not say what it thinks. It does not congratulate you for a good week or express concern about a bad one — it is not for either, and a companion that celebrates is also a companion that judges, just with the sign flipped.
The first few weeks this can read as underwhelming. You wanted a mirror that talks. What you have is a mirror.
Then at some point it says something like: Anil, four times this month. Not since February before that. And you sit with it, and you know exactly what it means, and nothing told you. That is the thing working.
If any of this resonated, you can read more about why self-awareness is not the same as self-criticism or the difference between venting and reflecting.
- Nisbett, R. E., & Wilson, T. D. (1977). Telling more than we can know: Verbal reports on mental processes. Psychological Review, 84(3), 231–259.
- Lieberman, M. D., Eisenberger, N. I., Crockett, M. J., Tom, S. M., Pfeifer, J. H., & Way, B. M. (2007). Putting feelings into words: Affect labeling disrupts amygdala activity in response to affective stimuli. Psychological Science, 18(5), 421–428.
- Pennebaker, J. W. (1997). Writing about emotional experiences as a therapeutic process. Psychological Science, 8(3), 162–166. The original study is Pennebaker, J. W., & Beall, S. K. (1986). Confronting a traumatic event: Toward an understanding of inhibition and disease. Journal of Abnormal Psychology, 95(3), 274–281.
- Bushman, B. J. (2002). Does venting anger feed or extinguish the flame? Catharsis, rumination, distraction, anger, and aggressive responding. Personality and Social Psychology Bulletin, 28(6), 724–731. Relevant as a caution that emotional expression is not automatically beneficial in the way folk theory assumes.