Excerpts from Article on Filk and Intellectual Property
As I've mentioned before, my co-authored article on Gender and attitudes toward intellectual property has been accepted for publication and will appear in early 2010. A draft is available for free public download at www.ssrn.com (search for the word "filk"). Because the full article is rather long, several folks have asked if I would post a summary - here goes:
Excerpts from Does Gender Influence Attitudes Toward Copyright in the Filk Community? by Melissa L. Tatum; Dr. Robert Spoo; and Benjamin Pope
Filk is not easily classified within the intellectual property laws in the United States. Since filk spans the entire gamut from original creations to lawful or unlawful derivative works, we wondered whether there was any difference between male and female filkers with respect to copyright. We were interested both in attitudes towards intellectual property and the types of songs written by members of each gender.
We built two databases - one consisting of songs themselves and one consisting of answers to a survey. We then analyzed those databases for relationships between gender and the other elements of the database. The first of our two databases, which contained 895 songs when complete, identified the author’s gender and whether the lyrics and/or the melody for each song used copyrighted material. The second database consists of the responses to a survey questionnaire administered to members of the filk community. The survey was admittedly not administered in a scientifically calculated method.
Highlights from the Song Database. The rates of publication appear highly connected with gender. Although men and women initially published songs at roughly the same rate twenty years ago, men quickly began outpacing women in the rate of publication, and indeed, between the late 1990’s and early 2000’s, there was a sharp decline in the number of songs published by women. Despite the fact that men publish songs at a greater rate than women, there was strong evidence that females are more likely to be nominated for a Pegasus Award. It is interesting to speculate about the disparity in nominations (particularly in light of the rates of publication). With the exception of the 2003-2007 time span, men and women were nominated for Pegasus Awards in very similar raw numbers. (It is not entirely clear why the 2003-2007 time period is so very different from prior time periods. We have shared our results with the filk community both online and at conventions, and two common explanations have been proffered. The first has to do with the rise of various technologies for distributing songs (such as the internet and mp3s), giving those who are likely to be at the cutting edge of such technologies (usually males) an edge in “growing” the audience for their work. The second is a change in nomination procedures for the Pegasus awards. We think it likely that both explanations are relevant.) And the raw numbers for multiple nominations is also very similar - 19 females and 20 males were nominated multiple times. This could simply reflect either a conscious or subconscious effort to distribute nominations equally among men and women.
We can’t help but speculate, however, whether the disparity is at least to some degree influenced by the originality of the lyrics and the melody. The statistics show that the type of melody used is strongly dependent on gender. Females were more likely to use completely original melodies, while males were more likely to use a copyrighted melody or the melody of another filker . This statistic may also explain part of the disparity in publication rates. We chose Xenofilkia as the primary source of our song database because it has been in continuous publication for over twenty years with a consistent editor. The nature of its format, however, means that it tends to lean heavily towards songs that are “to the tune of” another melody; it is not set up to publish many songs with original melodies. Other songbooks, such as the GAFilk Songbook, have a format suited to and a preference for original melodies, so future studies should be more inclusive with respect to magazine format.
There is weak evidence that the originality of the lyrics used is dependent on gender, although the originality runs the opposite direction here. Males were slightly more likely to use completely original lyrics, while females were more slightly more likely to use lyrics based on another's characters or universe.
Highlights from the Filk Survey Database. For the filk survey database, we elicited four general categories of information from each responder: basic background information; the types of filk songs written, performed and/or recorded by the responder; the responder’s understanding of copyright and IP law; and the responder’s opinions about that law. For each type of information, we evaluated whether there were gender-dependent differences among the respondents.
Of the 62 responses we received, 33 (or 53%) were from females and 29 (47%) were from males. We inquired about age, length of time involved in filk, educational background, occupation, and nationality. With the exception of occupation, none of these variables was dependent on gender. Occupation, however, showed strong evidence of dependence on gender. Males were more likely to work in technical jobs.
The survey elicited narrative answers to several questions seeking information about the responders’ understanding of copyright and IP law. Among these categories, only whether all legal requirements were considered showed strong evidence of evidence on gender (males 17%, females 0%). We also explicitly asked how each responder defined “fair use.” Among these categories, only one - whether fair use includes not profiting from others' work - showed strong evidence of dependence on gender (females 63.6%, males 31.0%).
Our sampling of filk responses is too limited to derive any strong conclusions from the data. Nevertheless, certain tentative inferences are suggested by those data. Women in the filk community are more likely than men to create original melodies to accompany their lyrics, while women are only somewhat more likely to borrow from others’ lyrics than are men. Because filk is often viewed as an imitative culture, the tendency of women to depart from that ethos in creating their own melodies seems significant. It might reveal a heightened contextual sensitivity to the legal rights of others, a sensitivity also suggested by the greater number of female respondents who believed that filkers should give credit to the authors from whom they borrow, and should not record copyrighted work without permission . Similarly, female respondents were much more likely to define fair use as not profiting from others’ work, and somewhat more likely to define it as giving credit to the original author and making private as opposed to public use of a protected work.
Male filkers, on the other hand, offered fewer responses about how to define fair use, while being significantly more likely to assert that they consider all legal requirements before engaging in filk activities . That men showed a markedly greater tendency than women to borrow the melodies of others would seem to be consistent with the imitative, competitive, and satirical impulses behind filking, and may suggest that attitudes of men and women to exploiting these impulses differ in certain respects.
Male respondents, however, were slightly less likely than female respondents to make use of the lyrics of others, i.e., slightly more likely to create their own original lyrics. Some of this may depend, however, on what respondents meant by “original” lyrics. Men might have been suggesting, for example, that filk lyrics that borrow from others’ songs but transform those lyrics to some degree are “original” lyrics. Women might have been more reluctant to make that claim, particularly in light of their greater sensitivity to giving credit to other authors and their markedly greater tendency to define fair use as not profiting from others’ work.
Finally, because use of others’ work is regarded by the filk community as a distinct type of homage, common attitudes towards infringement, piracy, and other copyright prohibitions simply may not prevail among filkers, or may be significantly inflected. The
foregoing data must be viewed, at least to some extent, in the light of this communal ethos. While some filkers take a step out of the filk community by marketing recordings of their songs, it would appear that most filkers confine their activities to cons and other traditional filk settings. These traditional, insulated settings, along with the imitation-is-the-sincerest-form-of-flattery ethos prevalent among filkers, suggest that filkers have crafted a pastime that they regard, quite rightly perhaps, as essentially and institutionally a type of fair use. This might account, for example, for the generally low response rate to our specific questions regarding fair use.
To the best of our knowledge, this is the first empirical study examining the types of songs written by male and female filkers and surveying attitudes of those in the filk community toward intellectual property law. It will take some time to fully digest the results, and it may even be fool-hardy at this stage to attempt to draw too many conclusions. We think it more likely that the best use of our work is as a springboard for additional empirical work targeting those areas where we found gender-related differences. Those studies should seek both to confirm (or refute) our results and to gather more information about why and how those gender-related differences arose.
Excerpts from Does Gender Influence Attitudes Toward Copyright in the Filk Community? by Melissa L. Tatum; Dr. Robert Spoo; and Benjamin Pope
Filk is not easily classified within the intellectual property laws in the United States. Since filk spans the entire gamut from original creations to lawful or unlawful derivative works, we wondered whether there was any difference between male and female filkers with respect to copyright. We were interested both in attitudes towards intellectual property and the types of songs written by members of each gender.
We built two databases - one consisting of songs themselves and one consisting of answers to a survey. We then analyzed those databases for relationships between gender and the other elements of the database. The first of our two databases, which contained 895 songs when complete, identified the author’s gender and whether the lyrics and/or the melody for each song used copyrighted material. The second database consists of the responses to a survey questionnaire administered to members of the filk community. The survey was admittedly not administered in a scientifically calculated method.
Highlights from the Song Database. The rates of publication appear highly connected with gender. Although men and women initially published songs at roughly the same rate twenty years ago, men quickly began outpacing women in the rate of publication, and indeed, between the late 1990’s and early 2000’s, there was a sharp decline in the number of songs published by women. Despite the fact that men publish songs at a greater rate than women, there was strong evidence that females are more likely to be nominated for a Pegasus Award. It is interesting to speculate about the disparity in nominations (particularly in light of the rates of publication). With the exception of the 2003-2007 time span, men and women were nominated for Pegasus Awards in very similar raw numbers. (It is not entirely clear why the 2003-2007 time period is so very different from prior time periods. We have shared our results with the filk community both online and at conventions, and two common explanations have been proffered. The first has to do with the rise of various technologies for distributing songs (such as the internet and mp3s), giving those who are likely to be at the cutting edge of such technologies (usually males) an edge in “growing” the audience for their work. The second is a change in nomination procedures for the Pegasus awards. We think it likely that both explanations are relevant.) And the raw numbers for multiple nominations is also very similar - 19 females and 20 males were nominated multiple times. This could simply reflect either a conscious or subconscious effort to distribute nominations equally among men and women.
We can’t help but speculate, however, whether the disparity is at least to some degree influenced by the originality of the lyrics and the melody. The statistics show that the type of melody used is strongly dependent on gender. Females were more likely to use completely original melodies, while males were more likely to use a copyrighted melody or the melody of another filker . This statistic may also explain part of the disparity in publication rates. We chose Xenofilkia as the primary source of our song database because it has been in continuous publication for over twenty years with a consistent editor. The nature of its format, however, means that it tends to lean heavily towards songs that are “to the tune of” another melody; it is not set up to publish many songs with original melodies. Other songbooks, such as the GAFilk Songbook, have a format suited to and a preference for original melodies, so future studies should be more inclusive with respect to magazine format.
There is weak evidence that the originality of the lyrics used is dependent on gender, although the originality runs the opposite direction here. Males were slightly more likely to use completely original lyrics, while females were more slightly more likely to use lyrics based on another's characters or universe.
Highlights from the Filk Survey Database. For the filk survey database, we elicited four general categories of information from each responder: basic background information; the types of filk songs written, performed and/or recorded by the responder; the responder’s understanding of copyright and IP law; and the responder’s opinions about that law. For each type of information, we evaluated whether there were gender-dependent differences among the respondents.
Of the 62 responses we received, 33 (or 53%) were from females and 29 (47%) were from males. We inquired about age, length of time involved in filk, educational background, occupation, and nationality. With the exception of occupation, none of these variables was dependent on gender. Occupation, however, showed strong evidence of dependence on gender. Males were more likely to work in technical jobs.
The survey elicited narrative answers to several questions seeking information about the responders’ understanding of copyright and IP law. Among these categories, only whether all legal requirements were considered showed strong evidence of evidence on gender (males 17%, females 0%). We also explicitly asked how each responder defined “fair use.” Among these categories, only one - whether fair use includes not profiting from others' work - showed strong evidence of dependence on gender (females 63.6%, males 31.0%).
Our sampling of filk responses is too limited to derive any strong conclusions from the data. Nevertheless, certain tentative inferences are suggested by those data. Women in the filk community are more likely than men to create original melodies to accompany their lyrics, while women are only somewhat more likely to borrow from others’ lyrics than are men. Because filk is often viewed as an imitative culture, the tendency of women to depart from that ethos in creating their own melodies seems significant. It might reveal a heightened contextual sensitivity to the legal rights of others, a sensitivity also suggested by the greater number of female respondents who believed that filkers should give credit to the authors from whom they borrow, and should not record copyrighted work without permission . Similarly, female respondents were much more likely to define fair use as not profiting from others’ work, and somewhat more likely to define it as giving credit to the original author and making private as opposed to public use of a protected work.
Male filkers, on the other hand, offered fewer responses about how to define fair use, while being significantly more likely to assert that they consider all legal requirements before engaging in filk activities . That men showed a markedly greater tendency than women to borrow the melodies of others would seem to be consistent with the imitative, competitive, and satirical impulses behind filking, and may suggest that attitudes of men and women to exploiting these impulses differ in certain respects.
Male respondents, however, were slightly less likely than female respondents to make use of the lyrics of others, i.e., slightly more likely to create their own original lyrics. Some of this may depend, however, on what respondents meant by “original” lyrics. Men might have been suggesting, for example, that filk lyrics that borrow from others’ songs but transform those lyrics to some degree are “original” lyrics. Women might have been more reluctant to make that claim, particularly in light of their greater sensitivity to giving credit to other authors and their markedly greater tendency to define fair use as not profiting from others’ work.
Finally, because use of others’ work is regarded by the filk community as a distinct type of homage, common attitudes towards infringement, piracy, and other copyright prohibitions simply may not prevail among filkers, or may be significantly inflected. The
foregoing data must be viewed, at least to some extent, in the light of this communal ethos. While some filkers take a step out of the filk community by marketing recordings of their songs, it would appear that most filkers confine their activities to cons and other traditional filk settings. These traditional, insulated settings, along with the imitation-is-the-sincerest-form-of-flattery ethos prevalent among filkers, suggest that filkers have crafted a pastime that they regard, quite rightly perhaps, as essentially and institutionally a type of fair use. This might account, for example, for the generally low response rate to our specific questions regarding fair use.
To the best of our knowledge, this is the first empirical study examining the types of songs written by male and female filkers and surveying attitudes of those in the filk community toward intellectual property law. It will take some time to fully digest the results, and it may even be fool-hardy at this stage to attempt to draw too many conclusions. We think it more likely that the best use of our work is as a springboard for additional empirical work targeting those areas where we found gender-related differences. Those studies should seek both to confirm (or refute) our results and to gather more information about why and how those gender-related differences arose.