How A Musicologist Views Digital Sampling Issues


by Judith Greenberg Finell

 

New York Law Journal, May 22, 1992.

 

This is the second installment of a three-part article. Part I appeared in last Friday’s issue.

 

THE DEVELOPMENT of digital sampling technology has resulted in new challenges for forensics — in detection, application of the law and the definitions of originality and infringement. So far, most sampling claims seem to be initiated by recording companies, though music publishing companies should also want to protect their copyrights in the underlying music.

 

With sampling, many of the traditional standards for evaluating a copyright infringement claim do not apply. Access is obviously not an issue, since sampling establishes access to the sampled recording. Nor is copying an issue, since a sample is a copy, even if the original has been altered electronically.

 

Therefore, determining whether one recording has sampled another requires forensic methods different from those used in traditional copyright infringement. Many of the elements that are considered insignificant in conventional copyright infringement cases take on greater importance in detection of sampling. In non-sampling cases, one usually compares two songs’ main melodies (pitches and rhythms) to determine similarity, then harmonies (chords), structure and style. In a sampling case, the identity of these key elements is a given.

 

On the other hand, tempo is rarely important in determining classic copyright infringement. Where sampling is in question, however, a good way to detect it is to clock precisely the speed of each song’s related portion. If they are identical, and other factors sound identical, too, sampling is a strong possibility. If they are not the identical speed, and the second piece sounds distorted and in a slightly lower pitch, then sampling could still have occurred, but perhaps the sample was slowed down deliberately by the infringer.

 

ANOTHER CHANGE in the comparison process is in identifying the actual key of each piece. In classic copyright cases, whether two songs are in the same key is irrelevant to whether one is an infringement of the other. Where the issue is sampling, however, whether the related fragments of the two songs are in the same key bears directly on whether the second recording sampled the first.

 

A third factor in detecting sampling that does not exist in standard copyright cases is instrumentation. In a traditional copyright case, whether the allegedly infringing work uses the same instrumentation as the original is normally not relevant. In a sampling case, instrumentation takes on a new significance. One way to detect sampling is to determine if the supposed sampler has left a “footprint” that enables the second song to be traced to the first. These “footprints” include the underlying accompaniment parts. In one case, a sampler had taken a fairly common piano passage from another recording. He might have escaped unnoticed had he not also included the first syllable of the singer-songwriter’s next vocal passage from the original recording. This one syllable proved sampling.

 

ONCE SAMPLING has been established, the main questions asked by attorneys for both plaintiffs and defendants are, “How much was taken?” and “How important is the sampled fragment to the original and to the new piece?” At this point the attorneys are planning negotiation strategy and need to define the strength of their position. Attorneys for defendants and plaintiffs ask for quantitative and qualitative evaluations in order to determine a reasonable fee to compensate the proprietor of the sampled recording. In making such evaluations, the relevant questions include:

 

  • What amount was taken from the original composition? Is it de minimis?
  • What percentage of the original recording was sampled?
  • What percentage of the infringing recording consists of the sampled material?
  • How important is the sampled material both to the original recording and the infringement? For example, is it the “hook” (main signature melody) in either piece?
  • Has the sampled material been changed or enhanced? If so, is it still recognizable as being derived from the earlier recording?
  • Could the new recording maintain its identity without the sampled material?
  • Has additional material in the new song evolved from the sampled material, thus enlarging the degree of copying?

 

Often, as is typical of rap music, a short musical fragment from the original song occupies a large portion of a new piece because it has been looped (repeated constantly). Because of its repeated use, this fragment takes on greater significance in the infringing piece than it had in the original piece. Also, if the new piece is rap music, then it is mostly spoken, rather than sung. Therefore, the sampled music fragment might be the only melodic material in the new song, and may be the musical glue that binds the piece.

 

Next week: Disguised infringement.

 

Judith Greenberg Finell is a musicologist and the president of Judith Finell MusicServices Inc., a consulting firm in Scarsdale, a suburb of New York City. She has served as an expert witness and consultant in various litigations involving copyright infringement, and often advises advertising agencies on making musical choices to avoid litigation. She has also done work in the area of digital sampling.

 

Copyright 1992 New York Law Publishing Company.