What the F?
Illustration by Ayo Arogunmati
Before the term sampling was popularized, dubbing was innovated by producers such as Osbourne “King Tubby” Ruddock and Lee “Scratch” Perry. These two producers used existing reggae music to produce riddim tracks, which were then deejayed over. Dubbing has influenced not only reggae music but also rock, pop, electronic music, and hip hop music. Sampling, the act of incorporating an earlier sound into a new work, became popular in New York in the late 1970s with a parallel development with other foundational aspects of the hip hop movement such as emceeing, breakdancing, and graffiti. With limited legal constraints in place, sampling attained its height in the late 80s and early 90s before a landmark ruling would change the hip hop creative ecosystem. The case described the act of sampling as stealing. A discussion of the legal doctrine along with a published paper sheds light on the case that shaped the sampling market and its impact on creativity and the right to fair use of a copyrighted work.
Limited monopoly rights are granted to copyright owners under United States law to incentivize more creative activity. US law views copyright from a utilitarian perspective with a primary goal not to benefit authors but to increase the public good. Copyright law covers two types of protectable works in music—sound recordings (i.e. actual sounds captured on cassettes, MP3) and compositions (i.e. sheet music embodying a song to be later performed). Authors maintain exclusive rights over both types of copyright. As an author, establishing infringement requires three proofs, which are ownership, access to the original by the person copying your work, and substantial similarities between the two copies of work. However, the protection is not absolute, proven only when the infringement falls outside of the utilitarian goals of the law.
The action of clearing samples is a relatively new action by artists and record labels. Prior to the landmark case between Grand Upright Music v Warner Brothers Records, the first opinion on sampling and copyright law, securing clearances was unnecessary. The case was filed against Biz Markie’s song, Alone Again, using an “unlicensed” sampling of Gilbert O’ Sullivan’s song, Alone Again (Naturally). The judge called the act of sampling stealing, failing to consider the public good. This case scared both artists and record labels, encouraging them to clear all samples regardless of amount or nature of the use. The result of the judgment is believed to have led to financial barriers, causing limited creativity and comprised work by artists.
Another relevant case occurred in 2005 between Bridgeport Music v Dimension Films that reinforced the idea that sampling a sound recording is equivalent to copyright infringement. This case along with Biz Markie’s was a catalyst for the development of clearance companies whose business model was to identify “sampled music” and cause the sampler to pay for the right to use the record. The cost was small, averaging around $1,000 per sample cleared, but escalating higher over the coming years.
A constant theme of the preceding cases is infringement with less attention paid to fair use, a presumption that is not afforded to the defendants in each situation. The early thought leaders on copyright ownership found that granting unlimited rights may stifle creativity, so fair use doctrine allowed for liability-free use of another’s work that meets the utilitarian goals. To determine fair use, the courts provided four key determinants.
Purpose and character
Nature of the copyrighted work
Amount and substantiality of the portion used
Effect of the use upon the potential market for or value of the copyrighted work (Market Effect)
The four key determinants have been discussed in the paper titled “Sampling Increases Music Sales: An Empirical Copyright Study” by Mike Schuster et al., and would encourage further reading on each specific determinant and its implications on fair use.
The courts deemed the market effect on the copyrighted work to be the most important factor in the fair use analysis, considering whether the copyright owner’s ability to exploit their work is diminished in the marketplace.
Because any use, other than the copyright owner, can be considered unfavorable to the owner, future cases were analyzed based on the question asking whether there is a current market for the composition. If there is an existing market, then fair use is voided by the courts. The problem that many of these future cases ran into was that a licensing market had been generated out of the wrong judgments from the Biz Markie case in 1991 and Bridgeport Music. As the licensing market for sampling increased as a result of the apprehension by artists and record labels to use and not pay for sampling, the application of fair use receded - what is described as doctrinal feedback.
For a licensing market that continues to expand, future samplers are more likely to pay for samples and reduce the application of fair use, which only encourages the cycle to continue. But research completed by Michael Shuster et al. highlights that there is a correlation between being sampled and additional sales increase of the sampled song. The analysis of the sales performed showed a 99% statistical significance in correlation beginning three weeks before the new song entered the charts. Furthermore, two to ten weeks after the new song was on the charts, the sales increase was 99.9% significant for the old sampled song. Based on the analysis, Schuster et al. concluded that the rise in purchasing is consistent with the hypothesis that incorporating the old song in the new song served as advertising of the sampled work.
Apart from just the financial implications, the authors looked at the specific attributes that resulted in the most increase in sales growth and found that increased sample sales correlated with the use of lyrical samples, different genres between the sampler/sampled, and pervasive use of the samples. It is unclear why the research on sales impact, which is the primary driver for the licensing market to begin has not been further developed, however, there is empirical evidence that suggests that the market effect is greater than the licensing income that is generated by each copyright owner.
The data engaged in the research showed a similar financial gain to the owner of the copyrighted material. In the early 90s, licensing revenue was on average $1,000 and $300 in legal fees, resulting in a net amount of $700. Fast forward to present day, an average sampled song realized an increase in sales of 2,106 units—at $0.60 per unit, this generates sampling advertising gain of $1,263.60, converted to 1991 dollar, the amount reduces to $703. This comparison cuts across both popular and unpopular samples and samples from the same or dissimilar genres, so further stratification to include only the samples from different genres led to an increase in the market effect income of $989 based on 1991 dollar.
Of course, there will be rebuttals to the argument to have free access to music to derive new music. The argument that customers may not know of the samples to start the discovery process is fair, but the new song can provide information to the customer on where the sound or lyric comes from. Another argument could be that judges may not recognize the fair use doctrine and conclude that a new song is stealing from the old song and destroying the market for the old song, however, the presumption of fair use can ease these concerns as artists and record labels will choose creativity over the potential for litigation for using a particular record. Another impact of fair use presumption is that it can enhance the advertising effect. More artists sampling means a greater number of copyright holders enjoying increases in sales.
A significant threat to creativity is the limiting nature of the licensing marketplace. A strategic approach by copyright owners is to allow free sampling of their songs under certain limitations. Schuster et al. suggest restrictions based on length or style including limitations to cross-genre sampling or by starting temporal limitations (i.e. use record only after a certain # of years in the marketplace).
With these changes, the market can move to an environment where the market effect supersedes the licensing charge or litigation charge. Many of our favorite records are based on a sample that introduce new ways of manipulating the old songs to fit with the present day. Limiting this creativity not only hurts new artists seeking to find their way but also impacts the financial return to the copyright owner. A more informed and researched process that uses data to provide artists and record companies with the correct information might move us in the direction where the use of sampling follows the utilitarian approach that promotes creativity for the public good.
 Schuster, W. Michael and Mitchell, David M. and Brown, Kenneth, Sampling Increases Music Sales: An Empirical Copyright Study (August 4, 2018). 56 American Business Law Journal 177 (2019, peer reviewed). Available at SSRN: https://ssrn.com/abstract=3226292
 If a [copyright] holder can show that it routinely issues licenses for a given use, then copyright law views that use as properly falling within the rightsholder’s control. Thus, the practice of licensing within gray areas [i.e., acts that are arguably a fair use] eventually makes those areas less gray, as the licensing itself becomes the proof that the entitlement covers the use.
[12-18] Schuster, W. Michael and Mitchell, David M. and Brown, Kenneth, Sampling Increases Music Sales: An Empirical Copyright Study (August 4, 2018). 56 American Business Law Journal 177 (2019, peer reviewed). Available at SSRN: https://ssrn.com/abstract=3226292