Header image

john hawks weblog

paleoanthropology, genetics and evolution

Photo Credit: Wikimedia Commons

Why is open science important in archaeology?

Are you curious about open science, but don’t really know what it means? The September issue of The SAA Archaeological Record includes an article that reviews “open science” approaches in archaeology: “Open Science in Archaeology”.

Marwick et al 2017 article header

This article was the brainchild of Ben Marwick, who has helped to organize the new Open Science Interest Group within the Society for American Archaeology. I’m proud to be able to support and participate in this group, and to have joined with 48 other professionals in this paper. I work at all levels within my scientific research to advance the principles that the article describes.

The paper begins with a short discussion of what “open science” actually means.

Often described as “open science,” these new norms include data stewardship instead of data ownership, transparency in the analysis process instead of secrecy, and public involvement instead of exclusion.

I approve strongly of this definition. Open science is nothing radically new, it reflects a recognition that responsible scientific approaches lie at one end of an axis, where the opposite end is really an antiscientific attitude of exclusion.

The new paper refers to open access (in publication), open data, and open methods. All these tend to increase transparency and replicability in the production of knowledge.

I would add a couple of aspects that the article doesn’t discuss in detail.

Archaeological sites are not merely data sources, they are physical places. Allowing colleagues and the public to see the sites and inspect work at sites is part of providing confidence and transparency in archaeologists as stewards of heritage. That access can be provided today with technology, as many projects (including our Rising Star project) are doing. Or access to sites can be provided in cooperation with national heritage authorities through responsible tourism and site visits.

Scientific projects are complex social undertakings that involve power and funding, and open collaboration may be just as important as open methods and open data in providing transparency of scientific processes.

Some people have the misconception that open approaches are less rigorous compared to approaches that involve long gestation of ideas in relative secrecy. Unfortunately, this misconception is still actively promoted by a few irresponsible scientists. Spreading such a misconception is much like the strategy of “fear, uncertainty, and doubt” that was once deployed by software companies in their battle for market share against open source software projects.

In my experience, open approaches are more rigorous than secretive ones. Open approaches rely strongly upon establishing transparent methods that emphasize replicability. When researchers follow through on a commitment to provide the data that underlie their analyses, they provide the means for independent researchers to check their results and conclusions. It’s a basic principle of scientific credibility: Conclusions that cannot be checked should not be believed.

The new paper is available and is a great resource. I know that academic articles about how to do academic work are not always exciting, but these articles are necessary to build the scholarly background for changing practices, especially in building support for responsible practices among institutions and grant agencies. I applaud the Society for American Archaeology for supporting this initiative.

There is no such thing as inertia—some people and institutions actively maintain processes that exclude colleagues and the public. Let’s subject those practices to examination and let institutions justify them if they are necessary. Meanwhile we must make the real costs of closed systems explicit, not hide them.

More: I’ve long been an advocate for open data practices, which I describe in my white paper, “Public interests in data from federally funded research”