All musicians who have worked with music on a computer have at some point encountered the legal issue surrounding the use of samples. Made possible by music technology, it is a copyright paradox that, on the one hand, one wants to protect expensive productions from abuse, but at the same time wants a diverse artistic freedom. A Google search on the subject is so misleading that Rumkraft has created a three-part article series that sheds light on the legal issues. The idea is that we will all become wiser on the subject so that we can better take a stand on it.

All musicians who have worked with music on a computer have at some point encountered the legal issue surrounding the use of samples. Made possible by music technology, it is a copyright paradox that, on the one hand, one wants to protect expensive productions from abuse, but at the same time wants a diverse artistic freedom. A Google search on the subject is so misleading that Rumkraft has created a three-part article series that sheds light on the legal issues. The idea is that we will all become wiser on the subject so that we can better take a stand on it.

The Phonogram Convention and the Producer Protection of Copyright Act

As mentioned in PART 1, a large number of countries decided back in 1971 that the manufacturer of an audio recording should be protected – the “source of investment”. To be the author, you have to create something original, but according to this very special rule (§ 66) you just have to make a sound recording, and then no one else can ” imitate ” it. This protection is called “producer protection”.

Aha? So what does that mean in practice? Well, if I go out into nature and record the sound of a horse squealing, then no one is technically allowed to copy the sound recording. The sound has (perhaps) economic value for a film company, and resources have been spent on producing it. Without Section 66, the recording would not be protected because the horse can never be the author. Another person may as well go out afterwards with a sound recorder and wait for 3 hours for the same horse to squeak. The word “imitate” covers only “technical copying” and not imitation, which was discussed in PART 2 .

But is it not technically possible to copy just a little? The wording “imitate” must – all other things being equal – mean that the sound recording must be recognizable before anyone can hear that it is an “imitation” – or what? In the phonogram convention, the wording is that one must “ duplicate” a “substantial part of the sounds” before there is a violation of the sound recording.

In the following, we will examine how Germany and the United States, respectively, have interpreted the phonogram convention. Unfortunately, there are no Nordic judgments that specifically take a position on sound sampling, although the possibility was there in the Les Djinns judgment. Typically enough for Danish courts, reference was simply made to sections 2 and 66 of the Act, without going into more detail on the limits of sampling. Even more typical is that the biggest copyright brains in Denmark cannot agree either. One, Morten Rosenmeier, is a professor at the University of Copenhagen. He believes that “an appropriate significant part” must be sampled before there is a violation. The other, Peter Schønning, is a lawyer for many of the rights holders in Denmark. He does not think at all that one can divide an audio recording into essential and insignificant parts. Therefore, sampling is in practice completely prohibited according to him.


Sampling in Germany

The German Supreme Court has said in the two “Metall auf Metall” judgments that all sampling is generally prohibited.

In short, the cases are about the music group “Kraftwerk”, which has worldwide cult status as pioneers in electronic music, published a musical work, “Metall auf Metall”, back in 1977. Two seconds of the work have been sampled and looped 20 years later. several times, by the female German rapper “Sabrina Setlur” under the song title “Nur mir”. It was thus necessary to assess whether this sampling was in breach of producer protection.

First of all, it should be mentioned that under German law it also requires originality if the musical work is to be protected by law. The court did not find that the 2 seconds met the originality requirement. In principle, therefore, the sampling did not infringe the work itself.

The system of German copyright law is very similar to the Danish one; therefore, the counterpart to § 66 (producer protection of sound recordings) was invoked by Kraftwerk. The court was then to decide whether the producer protection could be extended to protect the 2 seconds in question, or uphold the defendant in that, of course, it is just a matter of “inspiration” – what is called “free use” in legal language.

The court refused to place emphasis on either a qualitative or quantitative criterion, as an audio recording is not dependent on an artistic performance (originality). Thus, the court held that the producer protection is broader than that which applies to works under the normal system of law. The rule of “free use” could therefore not be used when it came to producer protection. In other words, the consideration of the cost and use of resources for sound recordings outweighs the free-use rule – and thus the artistic freedom.

In 2012, the case was reconsidered. Here, the German Supreme Court modified it so that the free-use rule can apply if the sound is too expensive to recreate for an average music producer.

The German Supreme Court has been criticized for not taking everything into account in the economic argument. Firstly, because music can never be recreated so that it sounds exactly the same way – not even by the same artist. Secondly, it does not make sense that sounds that can be reproduced relatively easily should not be sampled, whereas sounds that are difficult or expensive to reproduce may fall under the free-use rule.

Following criticism from many quarters, the German Constitutional Court on 31 May 2016 rejected the Supreme Court’s arguments, saying that artistic freedom is a constitutional right that outweighs the economic considerations behind producer protection. More about this in the section at the bottom, “Sampling today”.



Sampling in the USA

In the so-called Bridgeport I judgment of 2006, the court’s arguments were very reminiscent of the German Supreme Court. The ruling does not have as great an impact in the EU, as the legal system is structured fundamentally differently, but nevertheless the phonogram protection rules are similar, as all the countries that signed the phonogram convention have implemented the rules in much the same way.

The case was about sampling two seconds of guitar recording, which was otherwise processed by transposition, looping and extension to seven seconds, which touched on unrecognizability.

Under American law, an assessment is usually also made, reminiscent of the Danish one in imitation cases, but the court refused to use this assessment for audio recordings. In line with the German judgment, it was instead called for oneself to recreate unoriginal parts of a work. Although the “de minimis” principle (see below) had been recognized in an earlier US sampling decision, the court did not consider that there could be a de minimis threshold.

Thus, it was encouraged to create the sound itself, as 1) all sounds in reality have economic value, 2) the US licensing system already works well, as well as 3) a clear limit should be set for sound sampling.

This however has been criticized by that exactly will be many more lawsuits, since sampling is a reality that is not erased due. A single decision on the area. In addition, it results in a skewed balance of negotiations when the record companies can say “take it or leave it” The decision has also been criticized for contravening both the Congress’ intention that any copyright infringement requires the taking over of a “substantial portion” before liability can be established, as well as a Supreme Court decision, where i.a. ” Copying of constituent elements of the work that are original ” is necessary to establish infringement. Elsewhere, criticism has been that the decision undermines one of the most important purposes of copyright, which is to promote creativity.

In the USA, there is now a clear perception that you can burn the teddy bears very violently if you sample without the consent or license of the author and / or manufacturer. Therefore, if you are considering sampling something produced in England, Australia or the United States, you should think twice.


The exceptions

Law had been a nice easy discipline that one could program robots to learn – if it were not for the exceptions. If there is a main rule, you can be absolutely sure that there is an exception as well. In this article, I have chosen to mention only one exception. One could easily argue that the principles of “legal citation”, “freedom of speech” and “parodies” can justify sampling, but these are some long-haired discussions that continue to be enjoyed in my dissertation.



The law does not care about the smallest things

The illegal principle, “De minimis non curat lex”, means “the court is not interested in trifles” and can serve as a good argument in a given lawsuit about sampling. The principle is also expressed in certain places in the Copyright Act (§ 21). For example, you can go to the beach with your friends and play music on a mobile sound system, which is heard and danced to by many other beach guests, even though you would normally have to pay KODA for “public performance”.

The principle of de minimis can be used in other atypical, purely random contexts that do not directly follow from the law. But can sound sampling be justified by a purposeful interpretation of the de minimis principle?

The principle has in fact been recognized in a US judgment [1] dealing with sound sampling. The case concerned 6 seconds of direct sampling, which the Beastie Boys, however, had been given permission to use by the record company. But the composer’s rights had not been cleared. The court did not consider that the composition was original, but even if it had been, the de minimis principle would apply since the sound recording was sufficiently processed.

In line with this, Lubin [2] believes that the de minimis principle can serve as a balanced solution to the broad producer protection. He considers the court’s decision in the Bridgeport I judgment  to be correct, but to ban all digital sampling in the future does not agree with the premises of reality, where the processing through creativity should weigh heavier .

Unlike the traditional hip-hop sample, which is prominently featured in the newly created work, electronic music samples are used more as a tool and are largely unnoticed by the average listener. For this reason, the free-rider problem is avoided because the digital sampler does not experience the benefit of immediate recognition”

Following on from this, he argues that obtaining permission to use such small bites does not outweigh the massive, creative use of samples in electronic music. The limit for the de minimis principle must thus be set where the average listener experiences that the sampling has not only been used as a tool, but appears as the expression of the original work. If you want to take over such an expression, you must apply for permission.

If the de minimis principle is interpreted according to this standard in Denmark, one approaches the same infringement assessment that applies to works – ie the boundary between inspiration and imitation. However, the requirement for identity experience will be more easily met by direct sampling. The result will thus be that a given sampling almost always requires processing in order not to be “mood-creating”. Lesson III of the Black School will most likely not fall under the de minimis exception. Despite the massive processing and work effort of many months that has resulted in a completely different artistic expression in the form of cultural mergers, the directly acquired passages are too recognizable. Even though it is a completely new art form that carries on a cultural heritage that would otherwise be forgotten, it does not work.

In conclusion, it must be assumed that the de minimis principle cannot be ruled out as an argument for allowing unrecognizable sounds as a result of processed samples. In a consideration, the already vague economic reasons for producer protection seem to have been erased in view of 1) the widespread use of sound sampling, as well as 2) one of the main purposes of copyright, namely to encourage the creation of new works of art and literature .


Sampling today

The sampling debate is in rapid development. Every day, patent applications are filed in the field of music technology, and the inflexible law in the field is not geared to today’s rapid pace, where we are moving towards an AI era that can hopefully make infringement assessment more effective with intelligent sound wave analysis.

Until then, today we must take as our starting point “applicable law”. As mentioned above, on May 31, the German Constitutional Court ruled that artistic freedom outweighs producer protection. This development is not only important for Germany. The situation with copyright is such that it is an international cooperation, where the legal assessment in one EU country is important for all other countries in the European Community. It may even have an indirect impact on other Western societies that are not EU member states.

Should a case arise in Denmark, the courts are therefore obliged to follow the development in the other EU countries, including Germany, which as the only EU country has taken a qualified position on sampling. However, the German case is still not 100% settled, because the issue of sampling has now been sent to the European Court of Justice, and we are some who are jumping on the bandwagon to analyze the final decision, which will probably first come about a couple of years.

And now you are left with the feeling that it could be wonderfully liberating if you could get a definitive answer on where the limit for sampling should be set. First of all, one must remember that the outer limit of sampling is reviewed in PART 2 , namely the boundary between inspiration and imitation. In the section below, “The Pros and Cons”, there are some considerations that can be used when interpreting whether it can be justified to use a sound stump.

For example, if I have a saxophone solo that is technically difficult to play, and has otherwise undergone an expensive mastering process, the arguments would be that “production costs are high”, and that “performing musicians lose their livelihood if sampling is allowed” weighs quite heavily. Against this background, it is thus a good idea to either process the sound relatively much or just sample minimally. Try to weigh the considerations against each other yourself the next time you sample.


The pros and cons



  1. Production costs are high. It must pay to create a product.
  2. The independent right under section 66 of the Copyright Act provides an effective enforcement option for the producer.
  3. Small sound sequences have independent economic value, which can be sold in the form of sample packs.
  4. It is assumed in certain places in the literature that the performing musicians lose their livelihood if sampling is allowed.


  1. Sound sampling is a reality.
  2. It is almost impossible to detect processed samples with software. SoundCloud has tried with several errors as a result.
  3. The economic reasons that gave rise to the provision are largely blurred by modern music publishing and production.
  4. A massive rendering of a sound that results in a whole new expression cannot be associated with the original work in a copyright context.
  5. The exact same sound cannot be recreated.
  6. It seems extremely to break with the whole system of copyright when the legislature has not specifically taken a position on it.
  7. The de minimis principle can act as a sensible balancing act. If the new musical work cannot be said to be “the same” as the original work due to minimal sound sampling or massive processing, it does not infringe.


Sampling in the future

All indications are that we are heading towards a state of law in which sampling is recognized as a tool for creating new musical works. A development that can be described as natural when it comes to musical evolution.

If you are still a little shaken by the idea of ​​sampling, there is an Open Source alternative that makes sampling legal. Using works with a special license can easily become a big part of the sample-based music of the future. These are collectively called “CopyLeft” on the Internet and imply cc-konomarkthat you are welcome to edit the work, but if you want to publish it, it must be on the same terms as the original license, after which others can edit the work. Those of the licenses that may be used commercially usually contain a clause according to which one must credit the original artist. You thus never get an exclusive right to your work.

The popular Creative Commons licenses can also be used commercially without affecting KODA’s fees and charges. Koda is one of the first management companies in the world to enable its members to use Creative Commons licenses.


Still in doubt?

If the above is too abstract or you need legal advice within (electronic) music and sound, then finally send an email to







1) Newton v. Diamond 204 F. Supp. 2d 1244, 1248-1256 (C.D. Cal. 2002)

2) Lubin, Alex K. – Digital Music Sampling And Modern Music: The Need For The ‘De Minimis” Exception, Law School Student Scholarship. Paper 519, Seton Hall University.

Related Articles


Din e-mailadresse vil ikke blive publiceret. Krævede felter er markeret med *

This site uses Akismet to reduce spam. Learn how your comment data is processed.