Justice oriented science, open science, and replicable science are overlapping, but they are not the same

In an essay entitled “Open is cancelled” (January 14, 2020), A bee with a blog argued that “it’s time for open to disband”. The essay’s primary focus is recent events within the movement for free (libre) open source software (FLOSS), more specifically Jeffrey Epstein’s anonymous funding of the Media Lab at MIT and the statements that followed by Richard Stallman (the primary founder of the FLOSS movement during the early 80’s) and Professor Larry Lessig (founder of Creative Commons and former board member of Stallman’s Free Software Foundation). To learn more about these cases, you may want to start with the following two blog posts: “Remove Richard Stallman” and “On Joi and MIT”.

A bee with a blog (hereafter Bee) continues:

The leaders have proven themselves morally bankrupt. The community is toxic. […] It’s time to move on and create a new wave of ethics focused community management tools for code and content. […] It’s time to build a new movement, one fit for an era of rising fascism and climate justice. […] A movement that focuses on dismantling power structures and building solidarity across diverse groups. […] We need “justice oriented” data, “justice oriented” education, “justice oriented” science, “justice oriented” government, and “justice oriented” access to scholarly literature. […] The open movement failed when it centred freedom over justice.

Although Bee’s primary focus is on the FLOSS movement, the openness movements in general are discussed, including (open) science, (open) government, (open) education. As a researcher involved in the open science movement I will have a primary focus on open science, but similarly to Bee, I consider the topic and suggestions to be of high relevance also to the other openness movements.

The term “open” was introduced to create distance from a social movement: a brief history of software

FLOSS began during the late 1970’s and early 1980’s with Richard Stallman at an AI lab at MIT. “The free software movement […] is a movement for freedom and justice”, Stallman wrote. Although there may be somewhat different definitions of justice in this case, the Free Software Foundation (FSF) has since its beginning made it very clear that they are indeed a movement not only for freedom but also for social solidarity. “The free software movement campaigns for your freedom in your computing, as a matter of justice.”, FSF explains. “Does the issue of free software intersect with other issues such as gender, environment, or foreign policy?”, Stallman was asked. “Sometimes”, he replied. FSF’s strong emphasis on social solidarity and social justice was indeed what in 1998 made the free software movement split into two separate but overlapping movements: free (libre) software and open source software. The founders of the latter wanted a sole focus on the fact that free/open licenses strengthens the quality of the software. This group wanted to make it clear that they did not sign up to all the other philosophical and ethical aspects that the free (libre) software movement was fighting for. At a conference in Palo Alto, California, in 1998, the term “open source” was launched. From the Open Source Initiative’s website (history; bolding by me):

The conferees believed the pragmatic, business-case grounds that had motivated Netscape to release their code illustrated a valuable way to engage with potential software users and developers, and convince them to create and improve source code by participating in an engaged community. The conferees also believed that it would be useful to have a single label that identified this approach and distinguished it from the philosophically- and politically-focused label “free software.” Brainstorming for this new label eventually converged on the term “open source”, originally suggested by Christine Peterson.

Ever since the split in 1998 the Open Source Initiative (OSI) does not want to be associated with FSF and FSF does definitely not want to be confused with OSI. Stallman on Why “Free Software” is better than “Open Source” (bolding by me):

The Free Software movement and the Open Source movement are today separate movements with different views and goals, although we can and do work together on some practical projects. The fundamental difference between the two movements is in their values, their ways of looking at the world. For the Open Source movement, the issue of whether software should be open source is a practical question, not an ethical one. As one person put it, “Open source is a development methodology; free software is a social movement.” For the Open Source movement, non-free software is a suboptimal solution. For the Free Software movement, non-free software is a social problem and free software is the solution

FSF has developed terminology to further distance themselves from the open source movement. In a list of words to avoid we will for example find “closed” and “open”:

In the free software movement, we do not want to be confused with the open source camp, so we are careful to avoid saying things that would encourage people to lump us in with them. For instance, we avoid describing nonfree software as “closed.” We call it “nonfree” or “proprietary”.

[…]

Please avoid using the term “open” or “open source” […] Those terms refer to a different set of views based on different values. The free software movement campaigns for your freedom in your computing, as a matter of justice. The open source non-movement does not campaign for anything in this way. When referring to the open source views, it’s correct to use that name, but please do not use that term when talking about us, our software, or our views—that leads people to suppose our views are similar to theirs.

Ironically, it may have been FSF’s focus on social justice and solidarity that led to the cancelling of its own founder, Richard Stallman, in 2019. Posting on an internal MIT email list about the case of Jeffrey Epstein Stallman expressed himself in a way that could be interpreted as victim blame and was consequently pressured to resign from FSF. Similarly, Professor Larry Lessig received a wave of ugly protests after voicing concerns over the cancelling of an MIT lab leader who had accepted funds from Epstein. Unlike Stallman, Lessig remains in his position in the organisation he founded, but as Bee’s blog post shows, the protests do continue and on January 13th, Lessig sued New York Times for defamation.

On the adoption of “open” in science

Ever since the split in 1998, the movement for free (libre) open source software has been two separate movements, with neutral proponents and users in the middle (referred to as FOSS/FLOSS). But this fact was somehow lost in the transfer of the concept into other areas and disciplines. The term “open science” is today not only confused with calls for increased replicability and a more competent use of statistics but is increasingly often used to describe many of those philosophical and ethical aspects that actually was removed in 1998 with the introduction of the term “open”. Thus, today we have the term “open” that refers to one thing in software and at least three others things in the social and life sciences: “replicable research practices”, “transparent research practices”, and a social justice movement that advocates transparent and sometimes also replicable research practices. It seems to me that a confusion like this often occur when people are unaware of the history and try to reinvent the wheel. It has become a bit of a conceptual mess and I think it would be useful for us all to clean it up.

Clarification of the concepts

First things first. It would be great if we could all just stop referring to “open science” when we really mean “replicable science”. The only thing these two things have in common is that open science to some extent offer solutions to facilitate replicable science (e.g. public archiving of data and materials). Open science can be open without being replicable, and arguably, replicable science can at least to some extent be replicable without being open. My understanding is that the confusion of these two terms began in psychological science with the rise of the transparency and replicability revolution in psychology around 2010-2011. This confusion has already led and continues to lead to misunderstandings between academics and across disciplines. Academics who had been involved in the open access movement for several years were happily aboard the psychologists’ transparency wave but seemingly without realising that the psychologists primarily discussed how to increase replicability through the use of open practices. Meanwhile, many psychologists were riding the wave with little or no interest in open access. Similarly, mathematicians who had for generations been concerned over the use of statistics in the empirical sciences didn’t even get that there was a revolution happening because the empirical scientists, with the psychologists on the front line, referred to the revolution as “open science”. “Open science is also important”, a senior mathematician told me once, “but I have very little interest in that”. At that time the psychological movement for open and replicable science was still rather small and I still didn’t know enough to distinguish the terms. I was thus unable to convince him that the use of statistics and its role in replicability finally was about to become a thing in psychology. Arguably, the confusion of the terms “open science” and “replicable science” has hampered collaboration across disciplines and between individual academics because we all expect the other to understand what we refer to when using the term “open science”. Below is therefore an attempt to clarify each term’s basic meaning.

Open and replicable science

“Open science” refers primarily to a set of public archiving practices that have been made possible by technological advancement during the last two or three decades. These practices include digital and public archiving of data, materials, and reports. One might also want to refer to non-anonymous peer review as an open scientific practice.

“Replicable science” refers primarily to a set of methodological and statistical practices that minimise the risks of research not being possible to replicate. These practices include competent and insightful use of statistics; describing the purpose, methods and results in a transparent and honest manner; providing new or updated information post hoc (e.g. retroactive disclosure statements); conducting direct replications; collaborations on large-scale data collections; preregistration / public protocols (the former term is primarily used by social scientists while the latter has been used by clinical researchers for about two decades); registered reports.

Many researchers have an overlapping interest in these two topics. In such cases the most correct term seems to be “open and replicable science”.

Justice oriented science

During the last few years, however, more confusion has been added when an arguably new academic movement appeared: a movement based in identity politics (sometimes referred to as identitarian left) working for increased social justice in academic settings. The openness movement spread in academia in parallel with the growth of identity politics. Unsurprisingly, we soon had an overlap of these two movements due to dual interests in these topics by many individual researchers. However, we still to date have no new name for this new movement that merged openness and identity politics / social justice. Instead conflicts arose between camps: one camp claims that open and replicable science has nothing or little to do with diversity, inclusion, and social justice; one claims that openness is not open without a focus on diversity and inclusion; and a third camp is neutral. The conflicts sometimes get ugly, especially on Twitter, and I suspect this is pretty much how things went about before the 1998 split in the software movement.

Similarly to Bee I think we need a new term for this expanded way of thinking about open science as also including identity politics and social justice. This expanded philosophical and political focus is fairly close to that of the free (libre) software movement. Free (libre) science would therefore be one possible new term for this movement. It might not be the best though due to the heavy emphasis on freedom rather than justice and inclusion. As Bee wrote: The […] movement failed when it centred freedom over justice.” I am myself not a part of this particular movement and I am therefore not the best person to make suggestions, but Bee is and suggests ““justice oriented” data, “justice oriented” education, “justice oriented” science, “justice oriented” government, and “justice oriented” access to scholarly literature.”. Perhaps would thus “justice oriented science” (and e.g. “justice oriented software”) be a good description of the movement’s focus and be a useful term in inspiring further development of its goals and values.

Let’s face it: we are two different movements for openness/replicability

It is time to accept that we already are two separate movements: open/replicable scientists who want to include identity politics and social justice in the work for increased openness or replicability, and open/replicable scientists who don’t.

Individuals and organisations focusing on open and/or replicable science without being part of justice oriented science do not have to be proponents of diversity and inclusion, but they can be. An example of this from the open source software movement is that the Linux Foundation employs a Code of Conduct (CoC) that strongly promotes diversity and inclusion. It is thus possible to be committed to, and even heavily engaged in, work toward improved diversity and inclusion (or other topics such as climate) while at the same time preferring to keep these topics out of the work for open and replicable science.

The Free Software movement and the Open Source movement are today separate movements with different views and goals, although we can and do work together on some practical projects.”, Stallman wrote in Why “Free Software” is better than “Open Source”. I envision the same future for open and replicable science: we will continue to work together on projects where we have overlapping interests, but work separately on projects where a commitment to justice oriented science is needed. Not because we feel bad about each other or to avoid conflicts per se, but because we sometimes do have very different goals and methods.


Header photo by MoneyforCoffee from Pixabay

One comment

  1. Thank you. Just to build on your argument on why the Linux Foundation needed to adopt secondary documents like a code of conduct to prove it cares about human rights and the people in general, the reason is that human rights are not central in its foundational documents and ethos. Being focused on practicalities, the Linux Foundation is the “money oriented” version of the “justice oriented” FSF.

    From “Why open source misses the point” https://www.gnu.org/philosophy/open-source-misses-the-point.en.html :

    > These freedoms are vitally important. They are essential, not just for the individual users’ sake, but for society as a whole because they promote social solidarity—that is, sharing and cooperation. They become even more important as our culture and life activities are increasingly digitized. In a world of digital sounds, images, and words, free software becomes increasingly essential for freedom in general.

    Free/libre software is about the rights of people and this is made clear in its constitutional documents, the GPL and the GNU manifesto:
    https://www.gnu.org/gnu/manifesto.html
    https://www.gnu.org/licenses/gpl-3.0.html

    It’s therefore easier to apply policies like the GNU Kind Communications Guidelines https://www.gnu.org/philosophy/kind-communication.html , because they are based directly on the foundational philosophy of GNU.

    Copyleft, by design, can easily integrate and aid other democratic and grass-roots effort, for instance the anti-war movement, because wars are often based on lies and secrets. In the spirit of Justice Stephen Breyer’s argument in Active Liberty (https://en.wikipedia.org/wiki/Active_Liberty ), to address the usual red herring argument about how free software can be (ab)used to kill people, I recently wrote:

    > Let’s use a more concrete example (because it’s actionable, unlike fictional lawsuits in North Korea): are General Atomics or Lockheed Martin free to use GNU software within their weapons? Are we in agreement with it, even if they’re used to conduct extrajudicial killings abroad or to kill civilians?

    > And the answer is: yes of course I agree they can use GNU software, especially if they publish all the software used on their drones and war planes under GPL. Is this an embarrassing ethical conclusion? Not at all: if the software were under GPL, it would be more transparent about what kind of ethical decisions it makes. That would empower democratic processes to decide whether such weapons should be used or not. (It’s not democratic for the GNU hackers to decide whether a state should wage war on another, even if it’s tempting.)

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s