Many conflicts result from the way people interact with each other and with our planet1. Since 1992, a range of global initiatives have emerged to find a more sustainable and equitable solution to these conflicts. In 2015, the United Nations set a 15-year plan2,3, composed of 17 Sustainable Development Goals (SDGs) and 169 associated targets, to promote prosperity for all while protecting our planet3. These goals touch on all aspects of human life and therefore interact in complex ways. They do not exist in isolation, and synergies and conflicts can emerge from their interactions2,4–6. For example, traditional approaches to increasing agricultural productivity (SDG2) will lead to biodiversity and natural habitat loss affecting our ability to meet SDG15。

The interdependencies of SDGs were recognized from their inception4, but the effects of actions to achieve one goal on the ability to achieve others were anticipated only recently. Some work helped to highlight how interactions between SDG pairs can be negative or co-beneficial6,8,9. However, in many instances, the statistical approach limited the ability to make inferences relevant for interventions. One particular hurdle to date is the lack of recognition in these analytical approaches that interactions can and will vary depending on the socioeconomic characteristics of countries. Accordingly, there is little knowledge of the context of the network emerging from direct and indirect interactions between SDGs, and there are no robust inferences of associations between goals. This is important because efforts to meet SDGs in isolation can be counter-productive10 if they affect other SDGs negatively.

Not all interactions may be negative, and investment in some SDGs can have additional benefits on multiple goals. There are also other barriers to SDG implementation, as the status quo on some goals can be advantageous for some groups (vested interests) or indeed desirable (for socioeconomic reasons)11. Understanding the SDG network can help to find new indirect ways to progress on specific SDGs while avoiding non-SDG barriers11. Likewise, identifying indirect positive effects of SDGs on other goals can help define the best governance structures to capitalize on synergies and accelerate progress towards the 2030 targets12–14. Finally, using a robust and unified approach to estimating direct and indirect interactions among goals can help us determine whether those interactions differ among countries, which could also explain diverging views on SDG interlinkages.

人类与人类、与地球之间相互作用方式造成了很多冲突。自1992年以来，出现了一系列旨在为这些冲突找到更可持续和公平的解决办法的全球倡议。2015年，联合国制定了一个为期15年的计划，旨在在保护我们地球的同时促进繁荣发展，这个计划由17个可持续发展目标，169个相关指标组成。这些目标涉及人们生活的各个方面，因此他们之间的相互关系错综复杂。这些目标不是独立存在的，它们之间的相互作用关系既有协同作用也有权衡作用。例如，增加农业产量的传统方法会造成生物多样性和自然栖息地的消失，从而影响我们实现目标15的能力。

可持续发展目标之间的相互依赖关系在它们最开始的时候就得到了认识，但是实现一个目标的行动对实现其他目标的能力的影响最近才提出来。一些研究强调了可持续发展目标两两之间消极和积极的相互影响作用。然而，在许多情况下，统计方法限制了作出有关干涉的推论的能力。至今，一个特别的障碍就是这些分析方法没有认识到相互作用能够并且将会随着各个国家的社会经济特征而发生变化。相应地，从可持续发展目标之间的直接或间接的相互作用中产生的网络背景知识很少，并且目标之间没有可靠的关联推论。这一点很重要，因为孤立地实现可持续发展目标的努力，如果对其他可持续发展目标造成负面影响，可能会适得其反。

并不是所有的相互作用都是负面的，投资一些可持续发展目标可以为多个目标带来额外的好处。可持续发展目标的实施也有其他的障碍，因为一些目标的现状对某些群体是有利的（既得利益者），或者的确是令人向往的（出于社会经济原因）。理解可持续发展目标网络能够有助于发现在具体可持续发展目标上取得进展的新的间接方式，同时避免非可持续发展目标障碍。同样地，识别可持续发展目标对其他目标间接的积极作用能够有助于确定最佳的管理结构，利用协同效应加快2030年实现目标的进程。最后，使用可靠的和统一的方法来评估可持续发展目标之间直接的和间接的相互关系有助于我们发现它们之间的相互作用是否会随着国家而变化，这也能解释关于可持续发展目标之间相互关系的不同观点。