澳洲教授看安全科学的界限

近日我在Elsevier.com下载了一篇在《安全科学》（SafetyScience）待印发表的文章“安全科学的几个问题（Issues in safetyscience）”，该文章由澳大利亚国立大学社会科学学院Andrew Hopkins教授撰写，其中有一节讨论了“安全科学的边界问题（The boundaries ofsafety science）”。其实，这类问题在我国已经讨论了几十年了，目前还是没有权威的说法，还是理不清道不明。该文章假设由中国作者去撰写和投稿，我估计《安全科学》是不会接受发表的。

由于个人也是《安全科学》多年的业余义务审稿人，对安全科学的界限问题也有些思考和比较关注，因此摘译该节文章供大家参考或讨论。内容主要是意译，不一定准确。

--------------------------------------------------------------------

“作为《安全科学》学报的一个审稿人，在审稿中我经常问自己：这篇文章的主题真的是属于安全科学的吗？该文章是否适合该学报？例如，我最近审阅了一篇题为：“一个有安全结构和可以减小冲击波作用的密封风险的煤矿移动式避难舱”的文章。我回答编辑说：“我的看法是，本文是纯粹的工程问题，因此不适合该学报发表。”然而，这确实是一个政策问题，是编辑需要思考做出决策的。该学报的编辑随后写信给我说：“在我看来，该文章不适合刊登在《安全科学》。”但这个决定是正确的结果吗？这取决于我们所说的安全科学边界的界定。

该学报的宗旨声明：《安全科学》是服务于人类安全科学技术研究的国际媒介。它从人们的安全工作延伸到其他领域，如交通、休闲和家庭，以及人的危险活动所涉及的其他领域。《安全科学》是多学科的，它的贡献者和它的受众范围可从心理学家到化学工程师。该学报涵盖了安全物理和安全工程，社会、政策和组织方面，风险管理，安全控制技术的有效性，标准化、法规、检测、保险、成本核算方面，人的行为安全等。

如果采纳上述声明的说法为真实，则上述列举的对那篇文章的决定是错误的。但事实并不是这么简单，这涉及到一个把关的问题。《安全科学》是一个同行评审学报，需要有同行评审才能工作。如果事实无法找到安全科学同行的人提供必要的专业评价，则作为审稿人无法提交评价意见。此外，即使审稿人能够进行评价，他们不会关心属于同行外所关心的读者。按我的判断，上面提到的文章脱离了当前安全科学读者的领域；将该文章送到采矿工程杂志，编辑很容易找到审稿人，并且有更多的读者。

此外，《安全科学》学报刊登的内容必然影响安全科学本身的界限。是否属于安全科学的内容必须从它的从业活动进行推断。举例来说，气候变化是一个巨大的威胁着人类的安全问题，并且从词义上也是《安全科学》宗旨所涵盖的范围。但是气候科学不是安全科学的一部分。所有这一切都提出了什么是安全科学界限的质询。

我认为答案应该是务实的，而不是原则性的。安全科学同行是由来自他们自己能够识别的安全科学学校的人，他们出席同类安全会议的人，他们彼此间能够读懂有关安全出版物的人等等。这是一个看起来很混乱的定义，但它能够识别安全科学的流动和转移特征。如果采用网络分析可以更加清楚的确定同行。当我写下这些文字后，我看到《安全科学》最近接受发表了一篇题为：“在密闭容器中的甲烷/空气混合物爆炸参数的火花持续时间效应”。我判断该文章是超出安全同行的兴趣。显然，有其他审稿人和编辑与我持不同的看法，这篇文章的作者是来自北京理工大学爆炸科学与技术国家重点实验室。这似乎是中国研究人员正在测试安全的边界，并谋求加入我所说的安全科学界限。前面讨论的是关于安全科学主题的事情，它并不涉及安全科学是否是一门真正科学的问题。”

附原文：The boundaries of safety science

Andrew Hopkins

(Australian National University, School ofSociology, Canberra, ACT 0200, Australia)

Asa reviewer for the journal, SafetyScience, I frequently find myself asking: is the subject of this articlereally safety science? Is it suitable for this journal? For instance, I recentlyreviewed an article entitled: ‘‘A one-piece coal mine mobile refuge chamber withsafety structure and less risk of sealing under shock wave’’. I returned thepaper to the editor saying: ‘‘My view is that this paper is pure engineeringand therefore not appropriate for the journal. However this is really a matter ofpolicy so you might like to think about making a policy decision.’’ The journaleditor subsequently wrote to the author saying: ‘‘It seems to me not to beappropriate for publication in the journal’’. Was this the right outcome? Itdepends on what we mean by safety science.

Accordingto the editorial statement: SafetyScience serves as an international medium for research in the science andtechnology of human safety. It extends from safety of people at work to otherspheres, such as transport, leisure and home, as well as every other field ofman’s hazardous activities. Safety Science is multidisciplinary. Itscontributors and its audience range from psychologists to chemical engineers.The journal covers the physics and engineering of safety; its social, policyand organizational aspects; the management of risks; the effectiveness ofcontrol techniques for safety; standardization, legislation, inspection, insurance,costing aspects, human behavior and safety and the like. Taking this statementat face value, the decision referred to above was the wrong one. But it is notas simple as this. Safety Science is a peer-reviewed journal. This can onlywork if there is some relevant community of peers. If it proves impossible tofind people within the safety science community with the necessary expertise toact as reviewers, submissions cannot be assessed. Moreover, even if they can beassessed, they will not be read if they fall quite outside the areas of interestof this particular community. In other words the journal and its contents areinevitably and properly shaped by its readership and by its reviewers, not justby an abstract definition. My judgment was that the article mentioned above layoutside the areas of expertise and interest of the current readership of SafetyScience and that it would better sent to some journal of mining engineering,where editors will not have such difficulty finding peer reviewers and it ismore likely to be read. This position has far reaching implication s. Itinvolves the exercise of what has been called a ‘‘gate-keeping function’’. Moreover,influencing the content of the journal in this way necessarily influences theboundaries of safety science itself. The editorial statement above assumes thatsafety science can be defined independently of its practitioners. I believe, onthe contrary, that the content of safety science must be inferred from theactivities of its practitioners. This means that as the safety sciencecommunity evolves, so too will the subject. For instance, climate change is a massivethreat to human safety, and is theoretically encompassed by the editorialstatement. But climate science is not currently part of safety science,although one can easily imagine the safety science community embracing aspectsof climate science in the future, with the journal evolving accordingly. Allthis raises the question of what is meant by the safety science community.Again, I think the answer is pragmatic rather than principled. The safetyscience community consists of people who are associated with self-identifiedschools of safety science, who go to safety conferences, who read each other’ssafety-related publications, and so on. This is a messy definition, but it isone that recognizes the fluid and shifting nature of safety science. It would takea network analysis to identify the community with greater clarity. As I writethese words I see that Safety Science has recently accepted for publication anarticle entitled: ‘‘Effect of spark duration on explosion parameters ofmethane/air mixtures in closed vessels’’. I would have judged this to beoutside the current community of interest. Clearly other reviewers and editorstake a different view. The authors of the article come from the State KeyLaboratory of Explosion Science and Technology, Beijing Institute ofTechnology. It would seem that Chinese researchers are testing the currentboundaries and seeking to join what I have called the safety science community.The preceding discussion is about the subject matter of safety science. It doesnot deal with the question of whether or to what extent safety science is trulya science. That question will no doubt be addressed by other contributor s tothis issue.