How to Improve Your Creativity Through Reading: The Poet Who Made Borrowing an Art

In the Japanese imperial court of the early thirteenth century, a poet could be judged on how well he stole. The technique was called honkadori, which translates roughly as taking from an original poem. You took a line, an image or a whole situation from a famous older poem, and you built a new one on top of it. Done badly, this was theft. Done well, it was the highest thing a poet could do, and Fujiwara no Teika, who compiled the Shin Kokinshū around 1204, spent his life defining the difference.

The court where originality meant knowing everything already written

Before Teika, obvious borrowing from earlier poets was frowned on. He made it a discipline. His position was that honkadori must not be stealing, and that what separated a true allusive variation from plagiarism was the effect on the reader. A reader who knew the original would feel both poems at once, the old one sounding underneath the new one like a struck string, and the meaning would sit in the space between them.

Which produced a strange requirement. To write something new, a poet had to have read, and largely memorized, the entire tradition. The Kokinshū, the Man'yōshū, the poems inside The Tale of Genji. Nothing could be alluded to that the audience did not already carry. Creativity, in that court, was a function of how much you had read and how precisely you could reach for it.

Teika also compiled the Hyakunin Isshu, a hundred poems by a hundred poets, which centuries later became the basis of a card game still played in Japanese households at New Year. A man whose whole method was recombination produced the anthology that most Japanese people have partly memorized.

What a short story does to the mind that reads it

Psychologists have a measure called the need for cognitive closure. A person high in it wants an answer quickly, finds ambiguity unpleasant, seizes on the first plausible explanation and then freezes on it. That disposition is consistently linked to weaker creative performance, for reasons that are easy to feel: the first idea is rarely the interesting one.

In 2013, Maja Djikic, Keith Oatley and Mihnea Moldoveanu at the University of Toronto ran a simple experiment. One hundred participants read either a short story or a nonfiction essay, drawn from matched sets of eight, controlled for length, reading difficulty and interest. Afterward they completed a need-for-closure questionnaire.

The people who read the short story reported a measurable drop in their need for closure. The essay readers did not. The effect was strongest among people who read a lot in ordinary life, whether fiction or nonfiction.

The researchers' explanation is worth sitting with. Reading a story does not require you to decide anything. You can hold a character's reasoning without endorsing it. You can think alongside a person you would cross the street to avoid, follow their logic to its end, and put the book down without having committed to anything. That is practice in staying open, which is the state creative work requires and daily life rarely permits.

Treat this as one small experiment rather than a settled fact. A hundred participants, self-reported outcomes, a single sitting. A later study set out to replicate the finding and to test whether literary fiction also improved openness, cognitive complexity and divergent thinking, which is the kind of follow-up the original invited. The position to hold is that the direction is promising and the size of the effect in ordinary life remains unproven.

What the most inventive scientists do when they are not working

Robert Root-Bernstein and colleagues read through autobiographies, biographies and obituaries of every Nobel laureate in the sciences from 1901 to 2000, tallying their hobbies. They compared the results against members of the Royal Society and the US National Academy of Sciences, against ordinary working scientists in the Sigma Xi society, and against the general public.

The gradient was consistent. Ordinary scientists' hobbies looked like everybody else's. Members of the elite academies were roughly twice as likely to have craft hobbies. Nobel laureates were around three times as likely as ordinary scientists to have arts and crafts avocations, and roughly fifty percent more likely than academy members. The more unusual the pursuit, the sharper the gap: for glassblowing, woodworking, painting and sculpture, the differences ran far higher than for common hobbies like photography.

Nobel laureates in literature, meanwhile, were about three times as likely as the general public to be practicing visual artists and twenty times as likely to have acted. Rabindranath Tagore composed music for hundreds of his own poems. Gao Xingjian and Derek Walcott exhibited paintings professionally. Günter Grass was a sculptor and printmaker.

Root-Bernstein himself cautions against reading the numbers too hard. The comparison sets were gathered decades apart by different methods, and people with time for hobbies differ from people working two jobs in ways that have nothing to do with creativity. What survives the caution is the pattern that the scientists themselves reported: again and again, in their own accounts, the outside interest fed the inside work.

Reading is the cheapest way to widen the input

Set the three findings side by side and one picture forms. Teika's poets could only make new poems from poems they had absorbed. The Toronto experiment suggests that stories loosen the grip that stops a mind from holding several possibilities at once. The Nobel data suggests that people who produce unusual work tend to have unusual material in them, gathered from somewhere other than their own field.

Not everyone can take up glassblowing. Everyone can read something outside their own subject, which is the same move at a fraction of the cost, and the same reach across domains that helps you solve problems faster. A midwife reading about bridge collapses, an accountant reading about coral reefs, a novelist reading about supply chains. The value is in the distance.

How this applies today

Read deliberately outside your field. The habit that separated the laureates was cross-domain input. One book a month from a shelf you would normally walk past does more for your thinking than three more books from your own.

Keep a commonplace book. From the Renaissance to the nineteenth century, educated readers kept notebooks of copied passages, organized by theme, and returned to them for years. John Locke published a method for indexing one. It is the analog version of what Teika's poets did with memory, and it solves the real problem, which is that you forget almost everything you read.

Reread the things that struck you. Allusive variation depends on knowing a source deeply enough to bend it. A book read three times is more usable than three books read once.

Let a story stay unresolved. When you finish a novel and cannot decide what the ending meant, sit with the discomfort rather than looking up an explanation. That discomfort is the need for closure being exercised, and it is exactly the muscle the Toronto study was measuring.

Give children range rather than reading level. A child steered only toward books at their measured level is being kept inside one domain. Myths, machines, insects, a graphic novel about a country they have never heard of. Wide input now becomes recombination later.