The Making of an Urban Legend

In my youth I was taken in by a story current at that time. It involved collecting the filters from Marlboro cigarettes and peeling off the paper wrapper to expose some printed codes on the inside of the wrapper. The story I and my friends had heard, was that if you found a filter with a particular sequence of codes (I don't recall the exact sequence) and sent it in to Marlboro's headquarters, you would win a lot of money.

I happened to have a ready supply of Marlboro butts since most of the employees at the nursing home where I worked smoked Marlboros. As a non-smoker, I didn't find much pleasure in sifting among heaps of cigarette butts, but the prospect of the reward ahead spurred me on.

Eventually I found a butt with the proper code and sent it in to the company. A few weeks later instead of getting a check in the mail, I got a form letter denying the existence of such a contest and stating that I had been taken in by "unfounded rumors".

Ever since then I've been interested in these stories (now known as "urban legends"). Where do they come from? How is it that they appear almost overnight -- all of a sudden it is the hot topic of conversation at school, work, parties, etc. Then after a while, it dies away. The hardiest urban legends take on a life of their own and return -- like comets circling the sun -- following their own appointed cycle.

The official definition of an urban legend (from the alt.folklore.urban FAQ):

"An urban legend:

* appears mysteriously and spreads spontaneously in varying forms.

* contains elements of humor or horror (the horror often "punishes" someone who flouts society's conventions).

* makes good storytelling

* does NOT have to be false, although most are. ULs (Urban Legends) often have a basis in fact, but it's their life after-the-fact (particularly in reference to the second and third points) that gives them particular interest."

These stories propagate like viruses, moving from carrier to carrier. Epidemiologists try to understand why one virus causes an epidemic and another similar one only infects a few people. The same can be said about urban legends -- some of them stick around a long time, returning every few months, seemingly indestructible. Others make relatively little impact and die away. This brief article is an attempt to learn something of the nature of urban legends by examining the spread of one particular story.

Since the advent of email and news groups in the 1980s, urban legends have spread at an even faster rate than in previous years. With a few quick keystrokes any person can spread an urban legend to hundreds of even thousands of other people. There is even a news group devoted exclusively to the discussion and tracking of urban legends, "alt.folklore.urban".

More recently, the world wide web and its attendant spiders, robots, and search engines have finally enabled us to try to measure the scope and impact of an urban legend. A hot piece of forwarded email during the month of October 1996 was entitled US Standard Railroad Gauge or How MilSpecs Live Forever. (Click here to see a common variant. ) I had never heard this story, then in the space of a single week I encountered it 3 separate times. I mentioned it to a few other people, some of whom told me that they had just seen it a day or two before. This began to sound like an urban legend in formation.

Having honed my internet search skills over the last few years, I set out to find more about this story, and possibly contact the author. I had tried this tactic several times before without success. In the course of previous search attempts I had noticed the cyclic nature of some urban legends. I conjectured that if I could locate the origin of one such story, I could track its progress to try to verify the distribution cycles that I had casually observed with other urban legends.

This time I got lucky. My Alta Vista search yielded fewer than 10 hits. In looking at the dates associated with the hits, I zeroed in on the oldest copy which was dated May, 1995. It turned out to be a copy from a mailing list posting from March, 1994. There were several email addresses listed there (showing the history of who forwarded the story). I picked the innermost address and sent off my query to Bill Innanen. To my amazement a few hours later he replied that he had written the story and posted it to a mailing list in February, 1994.

I next ran a few Deja News searches and found many hits, most of them quite recent. Clearly this story was reaching a more active phase of its dispersion.

After a few minutes of investigation, I realized that Bill Innanen's story of "How Milspecs Live Forever" fits quite well into the definition of an urban legend. I realized that I would now have the opportunity I had desired to track an urban legend from its source.

Time for some data

I will discuss the web and news group searches separately, since they have different characteristics.

Web pages

Dates are not very reliable on a web page. Alta Vista and Hot Bot print a date next to the URL and size information. This date refers to the date in which the page was visited or indexed. The actual data found on the page could be either older or newer than the index date. Netscape displays a "last modified date" under the "View document info" option. This date is also not very reliable, since a simple modification to an existing page to fix a minor typo or formatting error will change the date. Despite these caveats, I found that dates on some web pages can be relied upon. These are generally web pages that contained archival material (such as mailing list archives or humor archives). Many of these are automatically generated and never touched again.

Many mailing lists archive their postings on the web. Such pages are valuable because the text of the posting usually contains a date that can be used to verify the origin of the posting. The May, 1995 posting I found probably indicated that the mailing list started its archive at that time, even though the particular message I saw had been originally posted in March, 1994.

Other examples of reliable dates are found in web pages that archive humorous stories. Several people have web pages full of jokes and urban legends. "Milspecs Live Forever" was in several such repositories from 1995 and early 1996.

Overall, I found this story in 34 web pages. Several were newsletters:

* Cause/Effect Magazine, Fall 95.

* Standardization Newsletter (3/96 and 4/96) (a military newsletter)

* Track Warrants - Electronic Internet Newsletter of Western Railroading (5/95, 7/96, 9/96)

* Southern California Scenic Railway Association (9/96)

Many of the rest were either humor or mailing list archives.

News groups

One of the big problems with news groups is that it is nearly impossible to keep up with all the postings in every group that one is interested in. It is easy to spend a few hours per day combing through several news groups without finding much of value, yet on certain occasions a thread or even a single posting will contain vital information.

I have been a heavy user of Deja News since its inception about 18 months ago. With a well-formed Deja News query, I have been able to quickly cut through the noise and find the articles I'm interested in.

The date posted is clearly recorded in the header of each news item. Once posted, the item is never changed. People post follow-up articles to the same news group that can be easily tracked because of similar subject lines.

Deja News came on-line about a year after "Milspecs Live Forever" was first written. It is possible that it was discussed a few times in news groups and not recorded in the Deja News archives. Based on the frequency of posting this story, I'm convinced that it was not a hot topic of discussion until late 1996.

I used Deja News to search for instances of this urban legend or follow-ups related to it. I was surprised by how many different groups this story was being discussed in. As could be expected, it was an active item in several news groups dealing with railroads and transportation. It was also being discussed in groups devoted to engineering, and of course, in alt.folklore.urban. Here is the data:

The vast majority of the articles have been posted during October 1996. Following the pattern as described above, this story went from relative obscurity to a hot topic seemingly overnight. Of the 67 news group postings, over 50 of them were during October 1996.

Variations in the contents of the story

Certain copies of the story are obviously derived from one variant or another. Interestingly, the copies on the news groups seem to have some differences from those on the web.

There seem to be at least 4 "strains" of the main story:

1. the oldest strain has a few distinctive features -- near the end of the story, the word "MilSpecs" is misspelled as "MisSpecs". Thanks to Bill Innanen's misspelling, it is pretty easy to locate exact duplicates of his original. In addition, the track dimensions are written out in a particular way -- "4 ft. 8 1/2 in (1.44m)". If these elements are present in the posting. then it seems likely that it is close to Bill's original distribution.

Most of the other variants correct the typo and change some minor part of the formatting, such as increasing the white space or revising the title.

Some of the news group postings changed the track dimensions to a more compact "4 ft 8.5 in". These postings also seem to have dropped the metric equivalent.

2. quite often the original posting is accompanied by several follow-ups -- early on, in February 1994, two follow-ups were added to the story. One starts, "I mentioned this to my wife, a medieval studies major ..." the other mentions Napoleon being slowed down by the ruts in Russia not conforming to the standard Roman gauge. Another talks about Lincoln's role in standardizing railroad gauges, another refers to a "seven foot gauge." Not all copies have all of these follow-ups, but these are the most common ones that have appeared together with the original. There have been numerous follow-ups responding to some aspect of the original or responding to one of the follow-ups.

3. a common variant has a new paragraph added at the end to try to make the story more humorous. It begins, "So, the next time ...". This was found only a few times on the web, but frequently in the news groups. I suspect that it is a recent change.

4. one strain is attributed to a professor in Texas named O'Hare. He seems to have forwarded the story to a widely read mailing list (I found no trace of his original posting). His name is attached to several news group copies (but no web copies). It seems that quite a few people thought that he was the originator of the posting.

The original story has been reformatted in many different ways. Sometimes people separate the word "Milspecs" into two separate words, "Mil Specs". Other times they rephrase the title or add editorial comments at the beginning and end. A European posting noted that the metrical conversion is incorrect, but left the text unchanged.

In two cases the story was totally rewritten -- in one instance to make it seem more scholarly and in the other to condense it into 2 paragraphs.

A few parts of the story can be viewed as "canonical":

* the actual dimensions, in feet and inches. This is a basic fact which all versions maintain.

* the question-answer format of the story. This style is very engaging. I've seen a few people cut it out (presumably to condense the story) but most people leave it untouched.

* the Romans, their chariots, and the ruts. This is one of the controversial parts of the story and has sparked the most response.

* the comments about milspecs and bureaucracies. Other than possibly changing the spelling, this is also left untouched.

What was the critical factor that caused this story to spread like wildfire?

Focusing only on news group and web postings of this story inevitably misses quite a bit of distribution data. For every public posting there were probably hundreds or thousands of private email messages containing the story. One person sent a public posting at the same time as addressing the story in private email to over 30 people (documented in the header to a news posting).

It is very likely that the story has also appeared in print -- I have tracked at least 4 newsletters that printed this story over the past year. These newsletters were all ones that had electronic versions on the web. There could easily be another 10 or 20 newsletters which printed the story but did not have a copy on the web.

The reliability of the web search engines also should be questioned. The original posting has been at Bill Innanen's home page for over 2 years, yet no search engine has managed to index it. How many more copies of the story are sitting in unindexed locations on the web is impossible to know.

The story was somewhat known prior to Bill's writing it down in February, 1994. Bill claims to have heard it first on a radio broadcast by Charles Osgood. One news group participant claimed to have heard this story off and on for the last 30 years.

Whatever the origin, by October, 1996 this story had full urban legend status. I suspect that the newsletters were important in circulating the story to wider audiences of people who then heavily forwarded the story.

From the limited data we have, here is my assessment of the distribution of this story:

* creation phase -- 2/94 - 5/95 -- circulation mostly via private email and mailing lists.

* expansion phase -- 5/95 - 9/96 -- publication in newsletters, humor archives and additional mailing lists. News group postings limited mostly to railroad enthusiasts.

* saturation phase -- 10/96 - 11/96 -- distribution in thousands of email messages and over 50 news group postings. Somehow the story crossed over into the internet mainstream. Starting on 10/4/96 it was discussed for a few days on alt.folklore.urban. From there it spread quite widely to many other news groups. It also had another round of discussion in the railroad oriented news groups.

* contraction phase -- 11/96 - ? -- postings in November, 1996 are declining in number. I suspect that by December the story will have petered out.

In private correspondence, Bill added some additional speculations regarding the first phase:

One additional piece of info for which I have only spotty conformation. In the slack period between when I wrote it and when it exploded, it appears of have been circulating at a low level in the professional military and military contractor communities. Their postings, lists and groups are generally restricted and thus are mostly invisible.

I corresponded with the first person to post about this story in alt.folklore.urban. He said that he first saw it in email from a co-worker. As to why it became a hot topic now, he had no idea.

I would like to posit that there is a certain threshold of distribution that must occur before a story can be propelled to urban legend status. Mass marketing people describe a certain level of market saturation that must occur before a product gets well known, and sales can take off. I think the same reasoning can be applied to urban legends.

Somewhere in late September, 1996, this story reached the magic threshold. From there it exploded in distribution, reaching urban legend status very quickly. It subsided back to a level of obscurity in November, 1996 as people stopped forwarding and discussing it. If we assume that for every web and news group posting of the story there were 1000 private email versions, we arrive at a figure of around 100,000 copies of the story in circulation.

What to expect from this urban legend -- when will it be back?

This story is missing elements that characterize some of the classic urban legends such as The Bricklayer's Accident Report or Craig Shergold needs your get well cards. Slapstick humor or pathos are big elements of those stories. I believe that is one reason that they come back over and over again. In general "Milspecs Live Forever" is somewhat too historical and scientific to catch on as a perennial urban legend. It also lacks the "gut issues" (money, disease, sex, violence) which seem to be staples of the most frequently recurring stories.

The story is fun to read and brings on a chuckle in most who read it. Therefore it fulfills the urban legend criterion of being entertaining (good storytelling).

Like my Marlboro story of so many years ago, people took this story quite seriously. Many of the news group participants hotly debated the authenticity of some or all of its assertions. An element of the recurring urban legend is verisimilitude -- it has to sound authentic, yet be very difficult to verify or refute. Just like the cigarette story, it is a little too easy to verify some of the facts in "Milspecs Live Forever". The most successful urban legends have a ring of truth, yet don't lend themselves to verification. For example, you can't look up "The Bricklayer's Accident Report" in a history book. [Note: J.H. Brunvand has traced this urban legend at least to 1918. See the chapter called "The Barrel of Bricks" in his book, "Curses, Broiled Again", New York: W.W. Norton, 1989, p.180ff]

Further research on urban legends

I will leave it to others (perhaps members of the Meme Research Group) to more accurately define the "email multiplier" which must be applied to determine the true impact of urban legends. Why bother studying and understanding the distribution of urban legends? I think they are a powerful means of communication -- if marketers could figure out how to tap into an urban legend, I'm sure they would. Some of today's crude spammers seem to be trying (albeit in a simple, transparent way) to grab on to this powerful means of communication. Imagine if you could pay to create an urban legend that promoted a product! This might eventually increase the level of skepticism and cynicism on the net, but the first few ones would probably be wildly successful.

Urban legends have not yet become a marketing tool (to my knowledge), but they have had an impact (mostly negative) on some businesses. The power of urban legends to annoy real life business has been well documented. Here are a few examples:

The Lexis/Nexis privacy scare galvanized many thousands of people over the summer of 1996. Just around the time it hit urban legend status, it started to appear in daily newspapers, radio and TV. Ironically the story had been true earlier in the year, but by June of 1996, the company had already bowed to pressure and revised the most offensive aspects of the database offerings. The privacy concerns are real, the story clearly struck a sensitive nerve in many readers. Even if Lexis/Nexis is no longer a privacy threat, I have a feeling that this urban legend will be back.

The Good Times virus scare needlessly frightened many thousands of people. This urban legend has been through its cycle quite a few times in the last few years. There is even a FAQ to debunk it. Think of how much time has been wasted by people in agonizing over this legend.

Intellectual property protection has been a major issue, especially among the lawyers for humorous authors such as Dave Barry. His stories are routinely passed off as urban legends. The copyright owners of his material have been making increasingly serious threats against online services and internet providers who (often unwittingly) act as vehicles for violating Dave Barry's copyrights. Having seen how frequently the text of "Milspecs Live Forever" was changed over the last 2 years, I can understand the anxiety of Dave Barry's lawyers when they see his work in thousands of web sites and news group postings.

Urban legends are an inescapable part of our cyberspace community. They are a powerful force that manages to reach thousands of people more effectively than many other forms of communication. Their humor usually brightens our day. They are occasionally viewed as a threat to someone's business interests. Further research into urban legends will help us better understand the phenomenon, and hopefully avoid (or at least contain) its negative effects.

A note on search engines

In order to do this paper, I inadvertently had to benchmark several of the leading web search engines. [Note: Deja News is by far the best news group search engine -- I used no others.]

I entered the exact same queries into several web search engines. In each case I used the syntax recommended by the engine's help system in order to match all query terms. I think the results say some interesting things about how much more room for development remains in this field.

The first 2 queries differ by only the "s" in Romans. The difference in results was quite dramatic.

The third query searches for the original typo in Bill's version.

The fourth query searches for part of Bill's original title.

Query:

1. gauge Romans ruts

2. gauge Roman ruts

3. Misspecs

4. "Milspecs Live Forever" (exact string match)

Results should be interpreted to mean "number of hits of the story/total number of hits returned by the query". Thus 10/13 means that of the 13 hits returned by the query, there were 10 instances containing some recognizable part of the story.