4.4.5.5 Limitations

Attempting to detect EOI in social systems can encounter additional difficulties compared to ecological and climate systems (Chapters 1.6 and 2.5). Careful thought is required when considering other positive tipping points in order to decide which system elements should be monitored and which could show these EOI, as they are likely to be system dependent. The EV transition example is occurring as a substitution; this contains a market shift and some amount of behavioural change (4.3.2), therefore we consider sales and EV adverts. Other positive tipping points will not necessarily have a behavioural aspect, or alternatively may almost exclusively exist as a behavioural and values change. These would require a different framing and would likely be constrained by data availability. These methods require high temporal resolution data which matches the relevant timescale of the system and is sufficient in extent to precede the tipping point. It is uncommon for this data to be available for social systems and careful consideration must be given for which state variable should (and can) be measured in social systems. Questions also remain about the timescales over which we could detect these changes in resilience and whether they would manifest early enough to offer a substantial lead time compared to other analysis methods, such as expert elicitation. 

It is also possible that this resilience loss framing is not consistent across all social systems. One key difference between social and ecological systems is the question of agency; it is possible that people are able to self-correct or that interested actors may try to strengthen the feedbacks which keep a system within an ‘undesirable’ regime, and that some abrupt transitions may be too rapid (or exogenously caused) to be detectable with EOI. Some social tipping points may have obvious alternate states, such as substitution of an incumbent technology for a new, low-carbon innovation, however this may not always be the case, especially when considering cultural and behavioural changes, and the drivers and likelihoods of these alternate states will differ across countries and cultures.