Is it OK to “steal” scientific literature and give it away for free?

Much of the knowledge that science produces is locked away behind paywalls. A piracy platform called Sci-Hub wants to change that: Through Sci-Hub, you can download virtually any study that has been published in a scientific journal, completely for free. Is it morally acceptable for Sci-Hub to do something bad (“stealing” scientific literature) in order to do something good (providing access to knowledge)? Yes — but probably not for the reasons you think.

Chances are you have heard about graphene, a fascinating material discovered in 2005 [1]. Graphene is quite amazing, because as a material, it is about as two-dimensional as is physically possible — it consists of a single layer of carbon atoms. Despite being extremely thin, graphene has a number of extraordinary properties, such as its tensile strength (you can pull at it much harder than at, say, steel before it breaks) [2].

Graphene sounds almost like science fiction, and the tale of its serendipitous discovery is nothing short of beautiful [3]. One of the areas where researches hope graphene will be useful in the future is (computer) electronics: Silicon-based transistors that are widely used in modern electronics are reaching limits in terms of miniaturization. An important innovation in this respect was the introduction of three-dimensional transistors a few years ago [4]. But even with the innovation that are three-dimensional transistors, the limits of mostly classical silicon-based semiconductors remain, and it might be necessary to switch to different material in the not too distant future. One such paradigm shift could come in the form of graphene-based transistors. In 2014, IBM has demonstrated the most impressive proof-of-concept so far [5], and recent research has demonstrated that it is indeed possible to engineer graphene in such a way that it behaves as a semiconductor [6].

Promising though graphene is for use in transistors, it might not be necessary to can silicon-based transistors just yet. In 2012, another amazing two-dimensional material was discovered: Silicene [7]. As the name implies, silicene consists of a single layer of silicon atoms. Just last year, in 2015, a first proof-of-concept of a silicene-based transistor was presented [8].

Are you intrigued by silicene and would like to read the papers that I have cited above? That’s no problem at all! That is, if you are willing to pay a total of 57 US Dollars for those two papers:

The price of knowledge? Just two scientific papers cost 57$.

Are you compelled by the science of silicene but not able or not willing to pay 57 US Dollars for those papers? Then you have another option: Go to the piracy site Sci-Hub (, enter each paper’s DOI (digital object identifier) or URL, and pretty much instantly receive the papers for free.

What has been possible for quite a long time with music, movies and video games is now possible with scientific literature as well: If you are willing to engage in the illegal act of copyright violation, you have pretty much all scientific knowledge that is preserved in scientific journals at your fingertips, almost instantly, trivially easily, and free of any cost.

Sci-Hub is quite clearly illegal. But is it also wrong?

How scientific publishing works (and why it doesn’t work well)

Science is a truly fascinating part of human civilization. It is integral to, broadly speaking, human cultural progress, and there is (still) a baseline agreement in most societies that the epistemic project that is science is worth supporting by taxpayer money. These days, it has even become rather hip to profess how much one “fucking loves science”. In short, science is, by broad consensus, a bedrock of modern human civilization— and yet, there is no really clear way to define science, neither conceptually nor functionally.

This fundamental vagueness that is characteristic of science as a whole is also characteristic of scientific publications. There is, probably, some baseline agreement that the scientific enterprise has to be documented in order to be rightfully regarded as a scientific enterprise (yes, this is somewhat recursive). The way science is documented is, for the most part, through written text: By translating the many and oftentimes convoluted thoughts and datapoints about the world into textual narratives that have a clear beginning and an end, science takes place. In that sense, scientific publications are not merely the “output” at the end of the scientific process, but they are an integral part and necessary condition of science. Or, in other words: If one does not communicate ideas about the world with the world in the form of publications, then science is not happening.

Publishing texts, then, is not only a vital part of science, but an essential one. However, the specific nature of scientific texts is not unambiguously and universally defined — there is actually no criterion whatsoever for what constitutes a scientific publication. Of course, there are generally accepted rules of thumb about what makes a scientific text scientific (“rigorous”, “clear”, “transparent”, “objective”, “referenced”, and other vague descriptors come to mind), but the whole practice of scientific publishing is simply one of convention. That is not necessarily a bad thing: One could adopt the stance that form follows function, and that the current multitude of scientific publication formats is simply the consequence of how science developed and of how science wants to “express itself”. The problem with such a belief is, somewhat obviously, that it is a quasi-teleological one: There is some natural, ideal way of how science communicates, and the current status quo of scientific publishing is a function of that ideal.

This teleological view is perhaps strongest when it comes to what people believe about peer-reviewed scientific journals. There is a romanticized belief, within science as well as among the science-minded parts of the public, that peer-review is an almost magical process: Scientific findings are submitted to critical, but fair scrutiny by perfectly objective and perfectly capable scientific peers. What passes peer-review is considered good science and, essentially, the truth; what fails to pass peer-review is bad science.

This idealized view of the peer-review process is demonstrably false and not even close to being true. Peer-reviewed scientific journals are not arbiters of objective truth, but rather club-like organizations that select publications based on highly subjective criteria. There is not only a fundamental subjectivity to the peer-review process, but also a systemic creation of dysfunctional incentives: Peer-reviewed journals generally suffer from a severe publication bias. So-called “positive results” (most often in the form of “statistically significant” results) have a much greater chance of being accepted for publication than “negative” results [9–12]. The subjectivity of peer-reviewed journals in combination with their biased incentives is one of the main reason for the current replication crisis in science. The way the current system of peer-reviewed journals operates rewards unfortunate and/or dodgy scientific practices such as passive ignorance of the garden of forking paths in data analysis [13] (most data analysis is not pre-specified, but rather performed contingent on the data at hand, which is a problem) or active “p-hacking” [14] (when working with frequentist statistics — something that is, epistemically, almost always irrational; but that’s a topic for another article — , you can basically always find some “statistically significant” parameters or differences of parameters if you look hard enough and exploit researcher degrees of freedom), and researchers are acting perfectly rationally when they respond to such incentives.

Even though the system of peer-reviewed journals does not work nearly as ideally as we would like to believe, journals are not useless. On a functional level, journals do something that is quite important: They curate and archive scientific publications. At first, this thought might sound somewhat silly: Today, the internet is everywhere, and you can publish texts easily wherever and however you wish — are you not reading an article on a nifty, modern, and easily accessible platform right now? Q.E.D. But if you give it a little bit of thought, it becomes quite clear that publishing scientific texts anywhere online can very quickly become a chaotic affair, smiply because the amount of scientific research that is published every year is humongous. The journal Nature is tracking publications in just 68 journals in its “Nature Index”. In 2015, those 68 journals alone have published over 57'000 studies [15]; given that there are thousands of journals, the total amount of studies published each year is almost certainly in the high six figure ballpark. Even though it’s difficult to vividly imagine what those huge numbers really mean, it’s easy to imagine why the service provided by the journals matters, a lot. Journals collect and curate the staggering amounts of scientific studies, an they preserve them for future retrieval. And, perhaps most important of all, journals help make all those studies discoverable. If all researchers published their work simply anywhere on the internet, the research would be online, but it would also be much more difficult to actually find anything, and it would be uncertain how long those documents would actually remain online. In that sense, scientific journals provide an extremely valuable venue: They are the places that collect and archive the huge amounts of scientific publications that are created every day of every week of every month of every year. Scientific journals are essential to the process of science.

Maintaining a scientific journal requires resources. It takes people and it takes time to perform the services that journals perform. In order to have people devote their time to scientific journals, journals partly rely on free labor: Usually, journal editors as well as reviewers work for free. But not everything about scientific journals can be done for free, so it’s also necessary to spend money on journals. This is a simple reality, and one that is not very controversial; of course you need money in order to keep a scientific journal running. However, the specific structure of the journal landscape is sometimes surprising to people who are not familiar with it. Most scientific journals are owned by for-profit corporations, and those corporations, consequently, operate their journals with the express goal of making a profit — a perfectly rational and just motive for a business, of course. But the market of scientific journals is not one where healthy competiton makes sure prices stay adequate. Instead, the market of scientific publishers mostly resembles a very dysfunctional one, an oligopoly [16], dominated by only five companies: Reed-Elsevier, Springer, Wiley-Blackwell, Taylor & Francis and Sage. In 2014, Reed-Elsevier, the biggest of the bunch, has generated around 3.21 billion € in revenue and around 985 million € in profits before taxes in 2014 [17] (that’s about 3.52 and 1.08 billion US Dollars).

The market for scientific journals is a multi-billion Dollar one. Where do those huge revenues come from? In part, it’s actually individual people and organizations who buy single studies— if you decided that the two studies about silicene mentioned above are worth paying for, you will generate a little bit of revenue for the companies that operate the journals in question. But the largest chunk of the revenues does not come from selling individual studies, but from selling bundled subscriptions of journals to university libraries. This might sound like a perfectly fine model, but one needs to remember that the market of scientific journals is a dysfunctinoal one — there is no competition, but an oligopoly [18]: Publishing companies force universities into costly “take-it-or-leave-it” contracts where subscriptions to journals are sold in bundles, and universities are essentially required to accept those contracts because the academic personnel needs to access some of the journals in the bundles. That is almost a prototypical form of price gouging (Ironically, the referenced article [18] is behind a paywall).

The classical model of scientific publishing results in a bizarre situation:

  • Public universities are forced to spend millions upon millions to renew their journal subscriptions.
  • However, the contract terms mostly mean that only university students and personnel have access to those journals.
  • If any other citizen, group or organization wishes to access studies in journals, they can either buy individual articles, possibly subscribe to the journal in question, or, if available, go to a university library and access the journals from there.

In the end, this peculiar system of scientific publishing means that only academic personnel and students have access to paywalled studies. Of course, most people can in principle somehow get their hands on studies if they invest time and money in order to travel to a university library that will provide them access. But more realistically, people outside of the academic sphere are faced with a simple choice: Pay up, or forget about it. And the possibly most bizarre fact about all of this: If an individual person actually ended up paying for a study, they will be paying for it twice — once as taxpayers who fund public universities, and once as individual citizens.

This classical, paywalled model of scientific publishing has been challenged in recent years by the advent of open access publishing. Many (oftentimes not-for-profit) journals have sprung up that operate with a completely different financial and ethical paradigm: All studies are completely freely accessible by anyone, and universities do not pay any subscriptions to journals. Instead, the open access journals are covering their costs on a per-article basis — if a study gets accepted for publication, the authors have to pay a fixed fee. The most famous open access journals are probably the Public Library of Science family of journals, better known under their “PLOS” moniker.

Open access publishing is on the rise; ever more studies are published openly accessible [19]. Even though the lion share of scientific publishing is still done with the classical, paywalled model, the growth in open access publishing is probably outpacing the overall growth in scientific publishing [20]. If this trend continues, then some day all published studies will be openly accessible, and the problem that Sci-Hub is addressing will cease to exist. Unfortunately, we are still a long way from there, and the problem of paywalled studies will stay with us in the foreseeable future.

Sci-Hub is liberating scientific knowledge by “stealing” it

The academic piracy site Sci-Hub is the work of a single person: Alexandra Elbakyan from Kazakhstan. The operating principle of Sci-Hub is fairly brillliant, no matter whether you agree with Sci-Hub in principle. Sci-Hub works in two steps [21, 22]:

  • A request on Sci-Hub will first trigger a search in the “LibGen” (“Library Genesis”) database. LigGen is a database that contains millions of copyright-protected pieces of literature, among them millions of scientific journals [23].
  • If Sci-Hub fails to find the requested paper in the LibGen database, it will access the paper directly from the journal by using passwords that academics have shared with Sci-Hub.
  • When a new paper is accessed in this manner, it will automatically be stored within the LibGen database.

These steps do not seem excessively complicated, but actually building an online application that can automate these tasks in a mostly flawless manner is impressive. The only current downside to Sci-Hub, from a user-friendliness perspective, is the fact that parts of Sci-Hub are operated in Russian. If you have basic command of Russian, you will have no trouble navigating your way. If you are not at all familiar with Russian, Sci-Hub will still be perfectly usable after a few clicks (only a few buttons that appear after you have searched for a paper are in Russian).

Sci-Hub’s logo, a raven carrying a key.

In order to form an opinion on Sci-Hub, it is not irrelevant what its motives are. More precisely: If Sci-Hub itself was a commercial operation aimed at generating money by pirating studies, that would be troubling. But that does not seem to be the case. There is no advertising on Sci-Hub of any sorts, and the costs of operating Sci-Hub seem to be fully covered by donations. As is explained on the site itself:

The project is supported by user donations. Imagine the world with free access to knowledge for everyone ‐ a world without any paywalls. Donate for this vision to become true. Make your contribution to the battle against copyright laws and information inequality. Even the smallest donation counts.

From the available information, Sci-Hub seems like a project that is completely motivated by idealism (or, if one had a more negative view of Sci-Hub, ideology), and there is no indication that Sci-Hub pursues any undisclosed nefarious goals. This means that Sci-Hub truly is what it says it is: A piracy site that knowingly violates copyright law in order to provide unfettered access to scientific studies.

“Stealing” vs. stealing

The internet, it seems, is inextricably linked to grey areas and outright illegal activities concerning copyright laws. The reason for that is obvious: Many forms of information that used to be bound to physical media, be it books, movies, music, video games, software, or scientific studies, can be transmitted purely digitally online. With the advent of broadband internet, getting the copyrighted content you wanted by means downloading or streaming has become much more comfortable and much faster than accessing it via physical media. We humans are, generally, cost-sensitive: If we want to consume some piece of copyrighted information, the amount of goods or services we need to trade for that information matters to us. The internet, of course, has ushered in a new period of piracy. If you are just a little bit savvy, you can get any kind of information in exchange for nothing. That is a very, very tempting value proposition.

Even though a number of online services, such as Netflix for shows and movies, Spotify and Google Music for music, and Steam for video games have demonstrated that customers are willing to pay for reliable internet-based distribution services of copyrighted material, piracy still exists, and it is likely to exist in the future. The interesting thing about piracy is that people who pirate are not at all willing to engage physical theft in order to obtain the desired content [24]: While a majority of people agree that it is wrong to steal, for example, a BluRay of a movie from a store, a subset of that majority would have few qualms about illegally downloading a digital file of the movie in question.

Stealing a movie from a store is generally perceived as actual stealing, but illegally downloading a movie is generally perceived as “stealing”, or not-quite-stealing. When you take a physical object without trading some goods or services in return, we have a clear zero-sum game at play: You have gained an object of value, but have failed to remunerate the other party to the amount of that value. By being better off, the other party is directly and unambiguously worse off. When you engage in not-quite-stealing, however, things get a little murky. You are clearly better off, because you receive the desired information (sans a physical object attached to it). But it is not quite easy to determine what is happening to the other party, because this situation is not a zero-sum game. By pirating, you have created a digital copy of some information, but you have not truly stolen anything; the original information is still fully present with the owner. But this cannot be a blanket argument in favor of pirating, because pirating does have an actual impact. By pirating some piece of information, you have actively avoided completing your expected part of a deal, which consists of remunerating the other party, generally the copyright holder. In real-world terms, when we pirate something, we create a very real loss of revenue to some party, because that revenue would have occurred had we completed the transaction legally. The usual response to this observation is that not every pirate would have legally purchased the content they pirated had they not been able to pirate it — “Yes, sure, i did download that movie illegally, but if I hadn’t been able to download it, I wouldn’t have gone to the store to buy it”. That argument is not completely without merit. Seeing some movie might be a preference we have, but a relatively minor one. It’s far too unimportant for us to actually spend money on it. But that preference is important enough to us that we will watch it for free, the only cost being the (probably very minor) moral concern about or social stigmatization of pirating.

There are only fairly weak arguments in defense of pirating cultural works. That doesn’t mean that any form of sharing of such content is to be banned — reasonably, it shouldn’t be, and in most Western countries, sharing content among family and friends is not illegal. But the category of pirating that is dominant today — downloading some sort of content you would like to enjoy, but are not willing to pay for — is not justifiable under any reasonable set of assumptions. If I wish to experience, in the broadest sense, some artistic content, but the creators of that content make it available only in exchange for some amount of money, there is no immediately plausible way to justify circumventing the creators’ stipulations. Pirating in the sense of one specific category of action (violating copyright by means of downloading content for free which the creators do not offer for free) is malum in se — it’s an action that is, normatively, categorically wrong, no matter the specific legal context. For example, in Switzerland, it is actually legal to download pictures, movies and text that protected by copyright, so long as the downloaded content is only intended for personal use. But just because it is legal to pirate content in Switzerland, that does not make pirating ethically right.

If the case for pirating copyrighted cultural works is, generally, weak, does that apply to scientific works as well? After all, copyrighted material is copyrighted material, no matter the nature of that material? No: Scientific studies are a distinct category of content, and one that is, by its very nature, governed by a logic very distinct from the one governing cultural works.

Why Sci-Hub is good, or: Why the logic of scientific truth claims requires openness

Sci-Hub offers several arguments in defense of its violation of copyright laws by means of pirating scientific papers. The first of the arguments is the following:

At this time the widest possible distribution of research papers, as well as of other scientific or educational sources, is artificially restricted by copyright laws. Such laws effectively slow down the development of science in human society.

This is, prima facie, a fairly compelling argument: Restrictive copyright laws mean that scientific knowledge does not diffuse readily, and these roadblocks ultimately result in science’s progress being slower than it potentially could be. This argument does not sound implausible. However, Sci-Hub fails to demonstrate in some empirical manner that this purported mechanism actually exists. If there is evidence to that end, Sci-Hub should present it, because that would constitute a very powerful argument. However, I suspect that producing any such evidence would prove quite hard —because I don’t believe that evidence to support this strong hypothesis really exists. On a purely theoretical level, proponents of restrictive copyright laws could make a compelling counter-hypothesis: Without strong copyright laws, there would be no system of more or less self-regulating scientific journals as curators of scientific knowledge, because the companies running the journals would have no assurances that their risk-taking and their investments would yield sustainable revenue.

Another of Sci-Hub’s arguments is the following one:

We advocate for cancellation of intellectual property, or copyright laws, for scientific and educational resources.

Here, Sci-Hub is openly calling for ending copyright laws for all scientific and educational resources, thus making all such content public domain content. I actually do not understand this argument at all: Everything that Sci-Hub is trying to do by enabling everyone access to scientific papers can actually be achieved under sensible copyright law. All scientific journals that operate under the paradigm of open access are doing so with clear and strong copyright law — very often, a variant of the Creative Commons copyright is applied. For everyday purposes, a Creative Commons copyright license can actually be of more practical use than the public domain status, because dedicating content to the public domain in one country does not make that work necessarily public domain in other countries as well [25].

Another of Sci-Hub’s arguments concerns the legality of “online libraries”:

Copyright laws render the operation of most online libraries illegal. Hence many people are deprived from knowledge, while at the same time allowing rightholders to have a huge benefits from this.

Here, I am drawing a blank. I don’t know what exactly is meant by “online library”, and I don’t know why copyright law should make most online libraries illegal, in principle. Again, if one thinks about open access journals that operate with a Creative Commons copyright licence, such as the well-known PLOS ONE, that systems seems to work perfectly fine — scientific papers are available freely and legally to everyone. The part of the argument that I agree with is that many people are, effectively, prevented from accessing scientific knowledge, though I perhaps wouldn’t use “deprived” in this context (but that is purely semantics).

The final argument by Sci-Hub is, again, one with very far-reaching consequences:

The copyright fosters increase of both informational and economical inequality.

I think that the first part of the argument, informational inequality, is perhaps a bit tautological: The very premise of the problem is that information is not available equally to everyone. But maybe Sci-Hub’s argument here is that informational inequality results in what is sometimes referred to as “cumulative inequality” [26]. Inequality in terms of access to scientific knowledge is not only manifest in one specific instance of attempted information retrieval. Instead, those groups that have access to scientific knowledge profit from an ongoing accumulation of knowledge, which, in turn, facilitates retrieval and processing of further knowledge, and so forth. Cumulative inequality is essentially simply the Matthew effect (“the rich get richer and the poor get poorer.”). But again, similar to the claim of copyright laws slowing the pace of scientific progress, evidence is needed in support of such a hypothesis. The same is true of the second argument in the above quote as well: If restrictive copyright of scientific papers leads to economic inequality — a hypothesis that does not sound implausible — , then it is necessary to present some evidence in favor of that claim.

Overall, the way Sci-Hub justifies its existence is not all that well argued for, I believe. The strongest potential arguments by Sci-Hub are, currently, lacking empirical evidence to back them up (or, put more charitably: the evidence possibly exists, but Sci-Hub has so far failed to reference it). In this view of things, Sci-Hub fails to make a strong case for Sci-Hub, curiously. The general problem with Sci-Hub’s reasoning, I believe, lies with the type of arguments that Sci-Hub engages in. For the most part, Sci-Hub is arguing in a functional manner: Strict copyright law of scientific studies leads to some grave consequences. That may or may not be the case — we need evidence to find that out. There is, however, a relevant non-functional argument in favor of pirating studies that sci-hub does not mention, and it that argument upon the very nature of scientific knowledge. Further above, I have talked about piracy and cultural works (movies, TV shows, music, etc.) and argued that there is no plausible justification to pirate such works. The same is not true for scientific studies. Stientific studies, the distilled results and arguments of scientific research, are a category of content that is very distinct of what is usually considered cultural works. It’s easiest to understand why in terms of speech act theory.

Both cultural works and scientific studies can be considered to be specific forms of communication. They are not just utterances, but rather, utterances that contain and convey a so-called “illocutionary” force [27, 28]: They do not only express the immediate content of the utterances themselves, but additionally, they make the claim that the conditions for the validity of what is being said are satisfied. But the specific validity claim that cultural works make differs categorically from the validity claim of scientific studies. Cultural works belong to the class of “expressive” speech acts: The speakers of such speech acts merely and at most claim that the content of the utterances in question are honest expressions of the speakers’ subjective, emotional state of mind. Scientific studies as utterances, on the other hand, make a much bolder claim: The speakers of such speech acts claim that the content of the utterances in question is objectively true. Such utterances belong to the class of “assertive”, or “propositional”, speech acts.

Expressive and propositional speech acts make claims about very distinct parts of the world — epressive speech acts refer to the truthfulness of our subjective mind, propositional speech acts refer to the veracity of objective facts. That is a very fudamental categorical difference: In contrast to expressive speech acts, propositional speech acts are universal speech acts. By uttering a propositional speech act, we are not merely expressing a purely subjective opinion, but we are claiming an objective fact, and by the nature of being an alleged objective fact, it pertains to everyone, not just to the speaker, or to some closed club. Such a distinction, of course, makes intuitive sense: When we listen to a song we like, we are aware that we are having a purely emotional experience. However, when we are reading about some finding in a study, it’s clear that what has been found in that study is claiming to describe some part of objective, universal reality.

There is no justification for universal access to expressive speech acts. We might find expressives enjoyable, shocking, beautiful or ugly, but there is no plausible way to argue that expressives have to be universally and freely accessible. The nature of such speech acts makes them fundamentally non-universal, and access to them can be fully controlled by the speaker. There are some special situations in which expressive speech acts become universally relevant, usually in situations in which the motivations for some action partly determine the consequences or sanctions for that action (for example, if someone accused of murder pleas for insanity, then it is of general, universal interest whether that claim is truthful). Mostly, however, the expressive is the domain of the non-universal.

There is, I believe, a justification for universal access to propositional speech acts in the case of scientific studies. The nature of propositional speech acts is such that they claim to convey some true information about the objective world. By the very act of making such a claim, the speech act in question automatically pertains universally to everyone. If, say, a study found that the kangaroo population in Australia was growing, that study makes a universal claim. The universality of the claim lies in the act of describing some part of objective reality: What is described in the study is claimed to be universally true — no matter whether you live in Australia or in Argentina, no matter whether you are an animal biologist or a bus driver, the findings of the study (the kangaroo population in Australia is growing) are true.

By now, it has probably beome clear that the idea of scientific studies as propositional speech acts claiming universal truth simply describes a fundamental principle of science. Of course, the whole point of science is to posit universal truth claims and to subject those truth claims to critical scrutiny. If a researcher published some finding, but did so with the explicit proviso that the finding is not universally true — i.e., that the findings are “true”, but not universally true and thus unreplicable — , that would not be considered science, but pseudoscience. Again, this principle is quite obvious, of course.

The problem with the current system of scientific publishing is that the principle of universality of truth claims is only partly implemented. Within the scientific system, the principle of universality of truth claims is generally accepted as an epistemic precondition of scientific work. However, practically, the principle of universality of truth claims is not realized: By keeping scientific studies behind paywalls, only small parts of society have access to the truth claims that are scientific studies. This goes against the nature of scientific studies: If scientific studies claim universal truth of their findings (they always do), then it has to be possible, in principle, to subject those studies to universal scrutiny.

For that reason, Sci-Hub is normatively good: Sci-Hub is helping to bring the practical reality of the dissemination of scientific studies closer to the conceptual requirements of examinability of universal truth claims.


The piracy site Sci-Hub is not just another regular piracy site, because the content they offer is categorically different from cultural works such as movies, TV shows or music. Ethically, pirating cultural works is mostly wrong. That is not the case with pirating scientific studies. By their very nature, scientific studies make universal truth claims, and it needs to be possible to universally scrutinize such truth claims. Restrictive copyright law in combination with paywalls makes this impossible. By pirating paywalled studies, Sci-Hub is doing something bad (violating copyright law) in order to do something good (make universal truth claims universally accessible) — and the good that is being done far outweighs the bad.


[1] Novoselov, K. S., A. K. Geim, S. V. Morozov, D. Jiang, Y. Zhang, S. V. Dubonos, I. V. Grigorieva, and A. A. Firsov. 2004. “Electric Field Effect in Atomically Thin Carbon Films.” Science 306 (5696): 666–69. doi:10.1126/science.1102896.

[2] Lee, Changgu, Xiaoding Wei, Jeffrey W. Kysar, and James Hone. 2008. “Measurement of the Elastic Properties and Intrinsic Strength of Monolayer Graphene.” Science 321 (5887): 385–88. doi:10.1126/science.1157996.

[3] Colapinto, John. 2014. “Material Question.” The New Yorker, December 22.

[4] Courtland, Rachel. 2013. “The Status of Moore’s Law: It’s Complicated.” October 28.

[5] Han, Shu-Jen, Alberto Valdes Garcia, Satoshi Oida, Keith A. Jenkins, and Wilfried Haensch. 2014. “Graphene Radio Frequency Receiver Integrated Circuit.” Nature Communications 5 (January): 3086. doi:10.1038/ncomms4086.

[6] Nevius, M. S., M. Conrad, F. Wang, A. Celis, M. N. Nair, A. Taleb-Ibrahimi, A. Tejeda, and E. H. Conrad. 2015. “Semiconducting Graphene from Highly Ordered Substrate Interactions.” Physical Review Letters 115 (13). doi:10.1103/PhysRevLett.115.136802.

[7] Vogt, Patrick, Paola De Padova, Claudio Quaresima, Jose Avila, Emmanouil Frantzeskakis, Maria Carmen Asensio, Andrea Resta, Bénédicte Ealet, and Guy Le Lay. 2012. “Silicene: Compelling Experimental Evidence for Graphenelike Two-Dimensional Silicon.” Physical Review Letters 108 (15): 155501. doi:10.1103/PhysRevLett.108.155501.

[8] Tao, Li, Eugenio Cinquanta, Daniele Chiappe, Carlo Grazianetti, Marco Fanciulli, Madan Dubey, Alessandro Molle, and Deji Akinwande. 2015. “Silicene Field-Effect Transistors Operating at Room Temperature.” Nature Nanotechnology 10 (3): 227–31. doi:10.1038/nnano.2014.325.

[9] Emerson GB, Warme WJ, Wolf FM, Heckman JD, Brand RA, and Leopold SS. 2010. “Testing for the Presence of Positive-Outcome Bias in Peer Review: A Randomized Controlled Trial.” Archives of Internal Medicine 170 (21): 1934–39. doi:10.1001/archinternmed.2010.406.

[10] Fanelli, Daniele. 2010. “‘Positive’ Results Increase Down the Hierarchy of the Sciences.” PLoS ONE 5 (4): e10068. doi:10.1371/journal.pone.0010068.

[11] Franco, Annie, Neil Malhotra, and Gabor Simonovits. 2014. “Publication Bias in the Social Sciences: Unlocking the File Drawer.” Science 345 (6203): 1502–5. doi:10.1126/science.1255484.

[12] Smith, Richard. 2006. “Peer Review: A Flawed Process at the Heart of Science and Journals.” Journal of the Royal Society of Medicine 99 (4): 178–82.

[13] Gelman, Andrew, and Eric Loken. 2013. “The Garden of Forking Paths: Why Multiple Comparisons Can Be a Problem, Even When There Is No ‘fishing Expedition’ or ‘p-Hacking’ and the Research Hypothesis Was Posited ahead of Time.”

[14] Simmons, Joseph P., Leif D. Nelson, and Uri Simonsohn. 2011. “False-Positive Psychology Undisclosed Flexibility in Data Collection and Analysis Allows Presenting Anything as Significant.” Psychological Science 22 (11): 1359–66. doi:10.1177/0956797611417632.

[15] “Global Overview.” 2015. Nature 522 (7556): S2–3. doi:10.1038/522S2a.

[16] Larivière, Vincent, Stefanie Haustein, and Philippe Mongeon. 2015. “The Oligopoly of Academic Publishers in the Digital Era.” PLOS ONE 10 (6): e0127502. doi:10.1371/journal.pone.0127502.

[17] “RELX Group Annual Reports and Financial Statements 2014.” 2015.

[18] Bohannon, John. 2014. “Secret Bundles of Profit.” Science 344 (6190): 1332–33. doi:10.1126/science.344.6190.1332.

[19] Laakso, Mikael, and Bo-Christer Björk. 2012. “Anatomy of Open Access Publishing: A Study of Longitudinal Development and Internal Structure.” BMC Medicine 10: 124. doi:10.1186/1741–7015–10–124.

[20] Laakso, Mikael, Patrik Welling, Helena Bukvova, Linus Nyman, Bo-Christer Björk, and Turid Hedlund. 2011. “The Development of Open Access Journal Publishing from 1993 to 2009.” PLOS ONE 6 (6): e20961. doi:10.1371/journal.pone.0020961.

[21] Oxenham, Simon. 2016. “Meet the Robin Hood of Science.” Big Think. February 9.

[22] Waddell, Kaveh. 2016. “The Research Pirates of the Dark Web.” The Atlantic, February 9.

[23] Cabanac, Guillaume. 2015. “Bibliogifts in LibGen? A Study of a Text-Sharing Platform Driven by Biblioleaks and Crowdsourcing.” Journal of the Association for Information Science and Technology, April, n/a-n/a. doi:10.1002/asi.23445.

[24] Balestrino, Alessandro. 2008. “It Is a Theft but Not a Crime.” European Journal of Political Economy 24 (2): 455–69. doi:10.1016/j.ejpoleco.2007.11.002.

[25] “Public Domain - Creative Commons.” 2016. Accessed March 7.

[26] Ferraro, Kenneth F., and Tetyana Pylypiv Shippee. 2009. “Aging and Cumulative Inequality: How Does Inequality Get Under the Skin?” The Gerontologist 49 (3): 333–43. doi:10.1093/geront/gnp034.

[27] Habermas, Jürgen. 1976. “Some Distinctions in Universal Pragmatics: A Working Paper.” Theory and Society 3 (2): 155–67.

[28] Searle, John R. 1976. “A Classification of Illocutionary Acts.” Language in Society 5 (1): 1–23.