“Deleted Scenes” are small pieces of historical research that, through no fault of their own, were cut from the published Bike Battles book, or never made it past research notes. But they live on as blog posts!
Ten speed? Boneshaker? PBL? Neighborhood greenway? Ordinary, cruiser, gravel grinder, safety, chopper, medium weight, velocipede, plush road, dandy horse, fatbike, balloon tire, fixie? Sidepath, bicycle boulevard, bike lane, cyclotrack, bicycle superhighway?
It’s a confusion of Babel: when we talk to each other about the past, of course we use the words that we know today. But sometimes those words meant very different things in the past than what we want to describe now. It’s one of the first lessons of history, this realization that we cannot fully trust the words that we use now to convey what we mean about the past. Students often start out by declaring that the meaning of a word is whatever the dictionary says it is; carved in stone. “According to Webster’s Dictionary,” they’ll write, “a bicycle is . . . ” But given a good enough dictionary, they’ll see that these reference works define words based on whatever usage is dominant at the time, and that words can have multiple, warring meanings that are explicitly contradictory.
This is especially evident when it comes to talking about bicycle history. In any given time period, there existed multiple names for different bicycle types, applications, and traffic accommodations. Some names persisted; others dropped from usage and are associated solely with one time period. Still others meant a specific thing in one era before disappearing and then being resurrected with a different meaning in a later time. In Bike Battles, I quote Luis A. Vivanco often. In his book, Reconsidering the Bicycle: An Anthropological Perspective on a New (Old) Thing, he writes “the question … what is a bicycle?…is closely related to the when of a bicycle, that is, its historical period and the diverse social and technical factors that influenced the shape and qualities of the object.” (25) The historically-shifting what of a bicycle is nowhere more obvious than in the changing uses of words to describe things related to the physical object.
Let’s start with an easy one. Bicycle is certainly the most common and commonly-recognized word for the item we generally know today. But it has certainly not been universal; in the first eras of two-wheeled human-powered transportation technologies in North America, they were referred to as “velocipedes” and “dandy horses” and “boneshakers.” Even the high-wheeled penny farthings, which most 21st century observers would see and call “bicycles,” were more commonly called an “ordinary” or simply “the wheel.” So, using the very clever New York Times Chronicle project to graph just the number of stories containing the word “bicycle” shows change over time, but does not necessarily capture all stories about the thing that we might call a bicycle. The graph does show how the bike boom of the 1890s was a media sensation; it’s hard to fathom the popularity of the bicycle in this craze until we see an illustration like this, where the number of articles makes later booms look minuscule. (But just look at that post-WWII nadir, and the slow build to the 21st century return of the bike.)
Drawing from a much larger universe of text than the NYT, a search on the Google NGram Viewer produces a graph with a number of noticeable similarities and differences.*
On this graph of the incidence of the term in the entire corpus of books scanned by Google, you can see the brief fad of velocipedes, as well as the decreasing need to specify “safety bicycle” to distinguish that new design from the earlier, unsafe ordinary. The line of “bicycle” usage shows the 1890s boom clearly, and a post-WWII decline followed by a late-20th century return. But there’s also something unexpected, something not as clearly visible in the NYT search — a significant increase in usage in the decades before WWII, with a peak in the war years. (I call the 1930s increase a boomlet in Bike Battles; see chapter 4 for more on the starlet boomlet and the wartime Victory Bike.)
Another NGram graph shows us some of the complexity of language — “bike” as an informal version of “bicycle” increases in popularity quite radically over time. What is it about the 1970s boom that liberates “bike” as an approved term? I’m not sure; is it related to a general informality of language and custom?
What’s not here also matters. Words that were used to describe 19th-century cycling are particularly difficult to analyze in this way — “ordinary” and “wheel” are insufficiently distinct in their meaning to give us any certainty that graphing their usage, either in the New York Times or in scanned Google Books, means that the word is being used in a book or article that is solely about the bicycle. They have multiple meanings, only some of which relate to cycling; even attempting to limit hits to usages of “wheel” to articles or books that also use the term bicycle fails to produce usable results, since bicycle was not universally used. Wheelmen, on the other hand, is sufficiently unique:
Turning back to the New York Times, usage of “wheelmen” seems to accurately chart the overwhelming turn-of-the-century popularity and power of the League of American Wheelmen. Really, it’s almost inconceivable to me that nearly one thousand stories in one year used the word “wheelmen” in a single newspaper — does that mean that there were an average of three articles each day referencing wheelmen in 1895? More plausibly, usage bounces along in the early 20th century as LAW membership faded, but local chapters remained. The final surprise on the graph is the small number of articles resurrecting the term “wheelmen” even as the LAW briefly returned under that name in the 1970s.
When we turn to the word “sidepath,” things get a bit crazy. In my previous Journal of Policy History article and in Bike Battles, I wanted to rescue the largely-forgotten history of the turn-of-the-century sidepaths, a diffuse political movement to create separate, bicycle-specific, hard-surfaced paths on the side of existing roadways, enabled by state law and paid for by a user fee system (and, very rarely and less successfully, through taxation). Much like the word sidewalk, sidepath not only referred to a specific physical design, but also a legal and fiscal system behind the physical object. But the word sidepath was also used in periods before and after the 1890s bike boom to refer to any alternate path to a desired end, sometimes physical and sometimes metaphorical. The Google Ngram captures several distinct trends — the 1890s sidepath movement clearly stands out from the background noise of other usage; particularly in the capitalized proper names of county Sidepath Commissions, state Sidepath Laws, and named, physical Sidepaths networks in various cities (mostly in upstate New York, but also Minnesota and elsewhere.)
In subsequent years, the use of “sidepath” to talk about bicycle-specific infrastructure died out — until it was suddenly resurrected, largely in the 1970s, to talk about the controversial things colloquially referred to as “mandatory sidepath laws,” even though the physical sidepaths of the 1890s no longer existed. These are two distinct things, named with the same word. It’s why I occasionally have bicycle advocates become unreasonably angry with my research; they spent parts of their lives committed to abolishing the widely-reviled “mandatory sidepath laws,” and conflate the thing that they’re angry about with the 1890s movement. (I wrote a whole post on BikeLaw.com trying to disentangle the meaning, with mixed success.)
Turning to a smaller sample size in the New York Times, we can see the transition in terminology for bicycle-specific transportation infrastructure. Sidepath clearly had its day at the turn of the century, while bikeway suddenly rose to prominence seventy years later, linked to the 1970s bike boom. Bikeway was an amazingly plastic word, referring to all kinds of different things, with some usages actually contradicting other uses of the identical term. But it’s possible that its plasticity explains its popularity; after the 80s/90s decline in practical cycling, the word has returned once again, even as we’ve added a long list of competing names for bicycle infrastructure.
One last graph, this time pointing out geography as well as time. While derailleur-equipped lightweight bikes existed in the United States through much of the 20th century, they were quite rare. That meant that when multi-speed lightweight bikes appeared in the US in larger numbers in the 1970s, the American populace needed a distinct name for them. Unlike Europeans, Americans needed to convey the meaning that a the new bike was a functionally and physically distinct object from what Americans considered a “bicycle” for much of the 20th century — generally a heavy (or mid-weight) single-speed, coaster brake-equipped bike for children. The comparatively-inexpensive, Japanese-component “10-speed” bike took cycling by storm in the United States, so much so that the name was used to refer to bikes with a variety of different available gearing combinations. It became a generic name for “adult bicycle with multiple gears, and drop bars for use on the road for recreational and competitive riding” — but primarily in America.
The sort of meta-analysis in this blog post has become a hallmark of what is known as the digital humanities; new ways of looking at language, literature, and history with digital-age tools. But I find it particularly applicable to the fractured history of cycling, where changing terminology and meaning seems endemic to the topic. And we’re not quite sure what we’re talking about even now. I was actually prompted to write this post — referring to research I’ve done over the last several years — after attending a public information meeting here in La Crosse, Wisconsin. The ostensible purpose of the meeting was to move forward on our planned “bike boulevards,” or low-speed urban streets with automobile and bicycle traffic mixed, but on which bicycle transport is prioritized and encouraged through a variety of design features. At this particular meeting, however, the phrase “bike boulevard” was being discouraged, replaced now by “neighborhood greenway,” which was described as having the exact same meaning but better political resonance for non-cyclists. Bike boulevard was now, at least in that 2015 meeting, a deprecated term of art. But it was the term used in city’s 2012 Bike/Ped Master Plan, and the term used in the most recent NACTO urban design guide. Forget about maintaining meaning over a century; we can’t even agree on a name for a thing that will last more than three years.
— James Longhurst, May 2015
Special thanks to Michael Egan, McMaster University, for introducing me to the endless rabbit-hole that is the Google NGram Viewer. Follow him at @EganHistory to see his occasional Google NGram finds and fantastic facial hair.
* A word of caution here about using either of these tools uncritically. It’s easy to make the conceptual leap from specific to general too quickly. It’s not in any way true that either of these tools definitively documents the usage, meaning, or changing popularity of these terms throughout culture. Instead, these are samples of literate populations, each with a host of caveats. In fact, I prefer the much smaller New York Times corpus because it is more difficult to claim its results as universal; instead, we know quite clearly that we’re talking about one particular northeast urban newspaper; its authors, editors, and readership are clearly delineated, and by definition the written articles therein are explicitly linked to dates. The Google Books scanning project, on the other hand, is a jumble; a decades-long project to scan the contents of major university libraries, the initiative has run into complicated legal battles and negotiations which have limited both the progress of the project and the materials which have been added to the corpus. On top of that, the metadata for the scanned works is riddled with complicated errors, as Google has (predictably) gone their own way in cataloging the material, disregarding the coding systems of generations of librarians in favor of algorithmically-determined descriptive data. As such, it’s easy to search the full text of Google Books, but not at all easy to know if the date of publication is correct. Particularly with journals and serials, I’ve come across some very strange errors. It’s common to see that a 19th-century journal will be incorrectly listed as having been published at the later date when a university library bound the individual numbers together for long-term storage.