Sunday, May 6, 2012

Burning Man Statistics

After 3 years and 54,711 pictures, my preliminary research regarding nudity at the Burning Man festival is complete.  The background, method, metrics, conclusions, and future work are detailed below.


See the wiki page for Burning Man to learn about the festival itself.  This festival is of interest because it has several unique qualities:
  • It's a secluded city of around 50,000 regular people that lasts for 1 week.
  • The event is heavily photographed from all angles.
  • The event is clothing optional, and is in a climate which encourages clothing-optional use.
Also of interest is my post from when I started on this project.


To collect a large sample of photos, I used a script.  I ran the script from the start of the festival until the end of November.  The script would check the Flickr RSS feeds for specific tags (listed in my first post but changing the year).  The script would poll the RSS feeds at a minimum of 30 second intervals, and a maximum of 15 minute intervals, depending on how many pictures were collected in the previous iteration.  Any new pictures it found were downloaded.  The script timing did change between 2009 and 2010, but shouldn't have an impact.

One limitation that Flickr has is a maximum of 20 items in the RSS feed.  If someone uploads 30 photos all at once, the script would miss 10 of them.  (The same is true for 100 people uploading photos, the most recent 20 are reported.)  This works to my advantage because it ends up creating a random sample of the overall pool of photos.

After the pictures are downloaded, they are all evaluated for instances of nudity.  The number of people partaking in the nudity were tallied.  The categories that were chosen are listed below along with their general legality:
  • Male Nudity - Illegal in most areas.  Includes anything which shows the genitals (painted, shirt-cocking, etc.) 
  • Nude Women - Illegal in most areas.  Involves nudity of breasts and genitals and may include bodypaints.  (This includes 1 instance in 2009 where the genitals were exposed but not the breasts.)
  • Top-Free (no cover) - Illegal in most states (but not all).  Bare-breasted females.
  • Top-Free (paint) - Laws are mixed.  Females with painted breasts.
  • Top-Free (see-through) - Legally usually same as bare breasts.  Females wearing items over the breasts which still show the breasts.
  • Pasties - Generally considered legal but not always.  Females wearing items which stick on, or are glued on, with the intent of covering the nipple and areola.
I should note that it isn't an easy task to classify pictures.  Some are grainy or low quality, some strongly imply nudity but don't actually show it, and some are difficult to classify accurately.  I did the best that I could.  If any researcher wants to double-check my statistics, I still have the original pictures from each year.

Secondly, I excluded pictures which would unfairly skew the metrics, like those taken during the Critical Tits bike ride.  If counting nudists in a city, one would not want to include the World Naked Bike Ride participants as many don't normally go naked.  Only a couple pictures ended up being excluded.

Here are the raw metrics:

Year 2009 2010 2011
Sample 21142 15342 18327
Attendance 43435 51454 53963
Nude Men NA 73 59
Nude Women 35 13 14
Top-free (no cover) 120 97 134
Top-free (paint) 43 7 23
Top-free (see-through) 27 20 25
Pasties 41 32 56

Although the raw metrics don't paint an accurate trend.  The sample sizes are different, and need to be scaled to be equal.  The 2010 and 2011 metrics are scaled to match the 2009.  Below is the adjusted table with a couple additional metrics.

Year 2009 2010 2011
Nude Men NA 100.6 68.06
Nude Women 35 17.91 16.15
Top-free (no cover) 120 133.67 154.58
Top-free (paint) 43 9.65 26.53
Top-free (see-through) 27 27.56 28.84
Pasties 41 44.1 64.6

Sum Top-Free 190 170.88 209.95
Sum Top-Free (incl. Pasties) 231 214.98 274.56

Other factors to consider are:

Burning Man attendees often say there is a lot of nudity.  Based on the nudity appearing in pictures vs. the population, I'm seeing about 0.6% participate.  Informal counts (in 1999) put the number closer to 5%.  Nudists that I talked to at Sunsport also guesstimated the number around 5%, with no apparent change in the nudity rate over the last 10 years.

There is a big limitation with the metrics: I don't have the total number of people who appear in the pictures.  Thus, I cannot find an accurate percentage of attendees who are nude or top-free.  My 0.6% number above assumes everyone appeared in a picture, which is a big and likely untrue assumption.

Additionally, I don't know the duration that people are nude or top-free.  They could take off their clothes just for the picture, or they could go all week that way.

In order for an instance of nudity to be counted in this research, it has to undergo a series of events:
  • There must be a nude or top-free person at the event.
  • There must be a photographer nearby.  (In some places, photographers aren't allowed or are strongly discouraged, like Critical Tits, the Human Carcass Wash, and any sexual-themed areas.)
  • The photographer must take a picture.  Either by accident (where nudity is in the background) or with some sort of consent (if the main subject).  Sometimes photographers will take a picture even if there isn't consent, but it doesn't seem to happen frequently.
  • The photographer must share the picture on Flickr.  This means they are OK with showing others nude on a public website, possibly against their consent.  It also means the photographer is willing to share their pictures freely and have a Flickr account.


There seems to be a general downward trend for photographers capturing and sharing nudity.  In 2009, I flagged almost 600 pictures that had nudity.  In 2011, I flagged almost 350.  I'm not sure why this is, since I didn't track this metric very close.  It could have an impact on the conclusions below.  However, since the metrics count people instead of pictures, the picture count shouldn't matter.

Full-exposure is on a steady decline.  Both men and women are less willing to be fully naked in public.  It's possible that photographers are choosing not to capture/share it, but the remaining conclusions seem to prove otherwise.

Regular top-freedom is on a steady rise.  Women are increasingly willing to take their shirts off and go bare-chested, like a vast majority of men at the festival.

Pasties have grown in popularity.  Many may be part of a costume, but others seem to only be there to de-nudify.

Body painting seems to be less popular.  See-through clothing has remained very steady.

Top-freedom in general rose from 2009 to 2011.  (It wasn't a steady increase though, so it may fluctuate each year.)

It seems like conflicting information that top-freedom is rising, but full nudity is falling.  I believe that this indicates a sort of polarization. Top-equality, as an equal rights issue, is gaining support and gaining momentum.  Exposure of the genitals, on the other hand, seems to be more taboo in our current sexualized culture.

The increase in the use of pasties at Burning Man supports this argument.  The easiest way for women to remain covered is to wear a bikini top or similar clothing.  The majority do.  Some choose to show more skin, but add painful-to-remove pasties to remain "covered".  Pasties became popular in the 1920's simply to avoid breaking the law, and are still used today for that purpose.  There is no good reason to wear pasties at Burning Man since there is no law regarding nudity.  Many want to be topless while still being "covered", and resort to an extreme measure to make it happen.

If you have any conclusions from the data posted or other feedback, you're welcome to post a comment.

Future Work:

There hasn't been much academic interest in the "clothing optional city" of Burning Man.  Personally, I believe that Burning Man is a gauge for our culture's participation and acceptance of nudity.  I just wish I had better methods of measuring it.  Perhaps other researchers can attend the event, poll a random sample of participants, or collect more reliable data.

If there are any regular attendees who read this blog, I have an easy request.  Each year, set up a video camera on a busy street during a busy day and let it record for a straight couple of hours.  This way, a researcher can tally the total number of people that go by, along with their state of dress.  This would give a much more accurate statistic than I can generate based on pictures.  Plus, a rough baseline was already set for this in 1999.

Please let me know if you're a researcher who is doing related work.  I'm very interested in the results.

As a last resort, if nothing better comes along, I may repeat this experiment in a couple of years to see if the trends are still holding true.