Is the influence of classic writers on modern ones diminishing?

by Callum Hackett in For the Ears

Late last year, some researchers in the U.S. conducted a study that tried to determine the changing literary influences on writers, and the story seemed to be making the rounds on the news last week. The essential thesis is this: the influence of classic writers on our contemporaries is not what it used to be. Put in those terms, it sounds rather condemning, as though today’s writers are not reading up on their Chaucer and Milton and so are less culturally informed. That’s certainly how it’s been reported, but what did the researchers really mean? And is the claim even true?
The paper’s basic purpose was to conduct the first large-scale stylometric analysis of literature throughout different periods (i.e. an analysis of different writers’ linguistic styles), which they achieved thanks to the marvellous resources of Project Gutenberg. The most important facet of the study, with the greatest ramifications on the conclusions we can draw from it, is that they focused on content-free words, which they defined as: “the ‘syntactic glue’ of a language: they are words that carry little meaning on their own but form the bridge between words that convey meaning.” Such words include and, but, to, and of.
They narrowed their analysis to 537 authors, all of whom had at least 5 works on Project Gutenberg, though given the copyright laws that govern the texts they’re allowed to archive, they ranged from 1550-1952. Through some fancy linguistic equations and mathematical analysis, they were able to determine that the stylistic influences of preceding writers on their successors has diminished considerably in recent times. However, it’s extremely important to recognise that this concerns stylistic influence - the ands, buts, and ofs that a writer uses. This is how we can estimate the historical position of a line of Joyce and a line of Shakespeare without knowing exactly who we’re reading. It has little to do with the content of their works - their ideas, stories, and arguments - which bear no relation to their sentence structures. It’s quite misleading to even call the effects on style an “influence”, as a writer’s idiosyncratic use of prepositions and conjunctions is more likely a product of the natural evolution of language, driven by many complex social factors, than by the books they read.
Of course, their analysis presented a continuous range of texts spanning 400 years, but to make a statement about anything “modern”, it’s necessary to determine an arbitrary boundary between that and “old” texts. While you or I might intuitively draw this line around 1900, the authors explicitly stated their “modern” writers as extending from 1784, so the results cannot be said to demonstrate an effect that is unique to the writers of today and yesterday - this goes right back to the end of the 18th century. They did go on to make further subdivisions, but this graph from their article demonstrates a paucity of data in some areas despite the number of books they covered:

Here we see that the vast majority of analysis was concentrated on works from around 1830-1930. There were far fewer works from the decades straddling either side of that period, and an extremely small amount from before 1750.
So what exactly can we learn from this paper? Not much. They’re right to suggest that it provides a statistical basis for the common perception that there exist “literary styles of the time”, but these regard linguistic style, not ideological influence or familiarity with classic literature. Our contemporary writers are perfectly able to read as much classic fiction as their forebears while demonstrating smaller amounts of linguistic influence. The reasons for the diminished effect are also inconclusive - while it is most strongly implied that it has to do with reading habits and aesthetic movements, it’s conceivable that the effects of linguistic influence, particularly in English, have been greatly affected by globalisation - a factor not taken into account.
So don’t worry, the writers of today are not necessarily ignorant of their cultural heritage, it’s just journalists once again making tangential, unsubstantiated assumptions about a research article that the researchers gave them no cause for making.
—————
Hughes JM, Foti NJ, Krakauer DC, & Rockmore DN (2012). ‘Quantitative patterns of stylistic influence in the evolution of literature’ Proceedings of the National Academy of Sciences of the United States of America, 109 (20), 7682-6 PMID: 22547796... Read more »