To fulfill the purposes above, a definition should capture how we talk and what we value. It should also be clear, without overall adverse effects when implemented. We will discuss nine conditions that constitute the theoretical framework for analyzing and constructing a definition of “social sustainability”. This is not to imply that the list is exhaustive, and fulfilling the conditions should be seen as a desirable goal of a definition, rather than an absolute requirement. It is possible that no definition can satisfy all conditions. Moreover, satisfying a condition is a matter of degree. Definitions will likely satisfy the conditions to varying extents; and if so, we should use the definition that fulfills the most important conditions to the highest degree.⁷

1. Ordinary language

The first condition is the ordinary language condition.⁸ It is desirable for a definition of “social sustainability” to be consistent with how the term is ordinarily used.⁹ This includes how it is employed (although not necessarily defined) by social scientists, practitioners in urban development and, to some extent, people in general.¹⁰ 

For example, in Edinburgh, a study showed great health and lifespan disparities among people living along the same tramline.¹¹ In turn, this has led many to condemn cities like it for being fully or partially socially “unsustainable” (e.g., Dempsey et al. 2011: 291f). If these intuitive judgments survive scrutiny, our definition of “social sustainability” should be compatible with this usage of the concept. Barron and Gauntlett’s (2002) definition (described in the introduction), for instance, is reasonably in line with these judgments, since it seems that the community is not equitable and fails to provide a good life for everyone.

There are several reasons for having a definition correlate with ordinary usage. Satisfying this condition helps avoid misusage and misunderstanding. First, it is hard to relinquish the initial intuitive understanding of a term. Therefore, the term risks being misused if it fails to adhere to ordinary usage, even after being explicitly redefined. Second, in work on risk assessment, for instance, policymakers have been found to often misinterpret different sorts of risk analyses conducted by scientists and experts (e.g., Hansson 2018).¹² One explanation for this misinterpretation concerns the definitions of “risk.” Scientists and experts use a technical definition that differs from the ordinary understanding of the term.

Another fact, which supports the ordinary language condition, is that scientific jargon makes facilitating a non-hierarchical relationship between the public and experts harder (e.g., Dillard et al. 2009, Boström et al. 2015). Inaccessible terminology can have detrimental effects on all processes involving the public. Using a term that deviates from ordinary language makes it more difficult, or even impossible, to successfully collect required information and engage citizens. Hence, the processes risk being less democratic and, in this sense, less legitimate if they rely on a definition that fails to satisfy the ordinary language condition.

Ideally, a definition should also help explain our intuitive judgments concerning social sustainability. We should preferably be able to explain (1) our judgments about certain cases being prime examples of social sustainability or unsustainability, (2) our disagreement and uncertainty about whether a city or neighborhood is socially sustainable, and (3) our comparative judgments about social sustainability.

For example, Barron and Gauntlett’s (2002) definition could explain why most people judge Cape Town to be socially unsustainable, why it is less socially sustainable than the city of Gothenburg in Sweden, and why people might disagree about whether Gothenburg actually is socially sustainable. While Cape Town is a diverse city, it does not fulfill any criteria provided by the aforementioned definition, which accounts for the judgment that the city is unsustainable. Gothenburg fulfills the criteria much better than Cape Town, which explains the judgment that Gothenburg is more socially sustainable than Cape Town. However, Gothenburg has a lot of problems as well, which yields both optimistic and pessimistic opinions concerning what the future holds in terms of equity, well-being and other criteria. In other words, some individuals judge the city to be socially sustainable, while others disagree.

Good reasons exist for thinking that a definition should also explain the common core of our social sustainability judgments, sometimes called the “unity desideratum” (e.g., Bradley 2012: 395). First, we want to be able to provide arguments in favor of our views and against others’ opinions when it comes to urban development. To ensure a transparent process, we must have access to the fundamental building blocks of our judgments regarding social sustainability. We acquire this basis through explaining our beliefs about different cases. Of course, “justification” should not be conflated with “explanation.” For instance, even if the justification for our judgments about social sustainability is poor, the underlying explanations can still give us access to reasons we would otherwise lack.

Second, from a pedagogical standpoint, it is more desirable to present a definition that can explain a series of instances, rather than attempt to teach someone the entire body of cases or outcomes of a certain kind. This is partly because it facilitates discussion between people from different backgrounds, professions, and positions (e.g., Brülde 2003: 58).¹³ It is easier to formulate a more fundamental and accessible idea than to list a range of instances considered good and just, although the latter is quite common in the current debate (e.g., Ghahramanpouri et al. 2013). The point is that having a definition of ”social sustainability” that operates through fundamental principles enables us to apply the principles through generalizations about cases and through predictions.

In the discussion about how “social sustainability” should be defined, there are many definitions that arguably should be disregarded or modified because of their failure to meet the ordinary language criterion. Specifically, such definitions do not correlate with ordinary language use and/or do not offer a unified core. However, many definitions of “social sustainability” could be reduced to a common core. In particular, research overviews identify a range of related themes, such as justice, culture, democracy, integration (Weingaertner and Moberg 2014), well-being, life satisfaction, social cohesion (Ghahramanpouri et al. 2013), social equity, quality of life (Ghahramanpouri et al. 2013; Weingaertner and Moberg 2014), social inclusion, participation, and place identity (Gustavsson and Elander 2013, 2014). The common core here could be well-being, which is often perceived in philosophical and psychological literature as “life satisfaction” and “quality of life.”¹⁴ For instance, if social cohesion were bad for us, then we would likely not value it. Hence, the value of social cohesion seems reducible to that of well-being – of course, this is just one of many examples.

2. Coherence, precision, reliability, measurability, and simplicity

Policymakers and practitioners need a concept that can be systematically applied and that is sufficiently clear and precise. For instance, previous work on a collaborative research project including building entrepreneurs has shown that entrepreneurs have lost contracts because of the various interpretations of “social sustainability”. They have simply misunderstood what the client meant by the term.¹⁵ More generally, according to many of the management regimes, such as SMART, we need to be specific about our goals if we are to achieve them (e.g., Doran 1981).

Similarly, when politicians want to “sell” their vision to the public, there is democratic value in the public understanding what the politician wants to sell. Politicians probably also strive for a somewhat clear picture of their own visions. A reasonably clear definition helps us avoid what is sometimes called “policymakers’ escapism,” which hides the complexity underneath all the “sustainababble” (Engleman 2013: 3). This trap is easy to fall into when politicians, as well as the general public, lack a clear understanding of the term.

Thus, for people to communicate effectively and make good (and democratic) decisions, the concept must be fairly clear and precise. However, we are not saying that any definition that is precise is preferred over a definition that is less so (compare this to the criticism of different “new public management” regimes in Power 1997 and Bevan and Hood 2006). Instead, a definition should be clear and precise while also fulfilling the other conditions. If the definition cannot successfully do both, these shortcomings must be weighed against each other.

In addition to ordinary language, there are five more conditions of adequacy that relate to the considerations just discussed. First, we have the coherence condition, which simply states that a definition should be logical and consistent. Otherwise, it is difficult to even talk about social sustainability in a general manner. We are striving for a coherent set of defining features, such that a specific state is never both socially sustainable and not socially sustainable. In ordinary language, “social sustainability” might have a meaning that yields this sort of incoherent result. Our concepts are often vague and not thoroughly considered, even if they are in fact used in policy contexts. Consider the terms “health” and “disease.” These are not clearly defined in settings where policy is construed, and their meaning is presumably even less clear in ordinary language. If the ordinary usage of “social sustainability” is vague and incoherent in this way, we need to ask ourselves if coherence or adherence to ordinary language carries more weight. The goal should probably be to find a coherent definition as close to the ordinary language use as possible.

In a sense, the coherence condition is a precondition for the next two criteria, the precision condition and the reliability condition. According to the precision condition, a definition should be sufficiently clear and precise to, in principle, determine if a given state (e.g., outcome, process, city) is socially sustainable. In other words, there should be little doubt (ideally no doubt) as to whether a certain state is an instance of social sustainability. What it takes to satisfy the condition depends on whether the approach is categorical or dimensional. If we think of social sustainability as a category (i.e., determined by a certain (high) level of social sustainability), a definition should draw a sharp conceptual boundary between social sustainability and other desirable states, such as environmentally and economically valuable states. If we adopt a dimensional approach rather than (or in addition to) a categorical approach, the condition requires being able to determine whether an attitudinal (or other) change constitutes an improvement in social sustainability, a reduction in social sustainability, or neither.¹⁶
 

The reliability condition states that a definition should be practically applicable. It should be relatively easy, in practice (not just in principle), to determine whether a state is socially sustainable, or whether it is more or less socially valuable than another state. A concept that satisfies this condition enables different actors to apply the concept in the same way; for example, they might agree on what to exclude from the category of “social sustainability”. This condition is likely satisfied if the criteria for applying the concept are operational, which means the concept is defined in observational terms, and its presence can be established on observational grounds. Practical applicability is obviously important for communication purposes; a definition that satisfies this condition can (if commonly accepted) improve or facilitate communication between different groups and individuals, both within and across settings and cultures.

The next two conditions, the measurability condition and the simplicity condition, relate to each other, and to the greater question of assessing degrees of social sustainability in a transparent and systematic manner. According to the measurability condition, a definition should ideally make it possible to compare different levels of social sustainability, meaning that social sustainability is measurable in the ordinal sense. This holds for comparisons that are interpersonal, intrapersonal, and intertemporal.

However, a plausible concept of social sustainability likely does not allow for cardinal measurability, since some social values by which we measure social sustainability, such as attitudes, are not measurable in such a strong sense (e.g., Haybron 2016, Bykvist 2016). Furthermore, we believe that the measurability condition cannot be satisfied to a high degree unless the following simplicity condition is satisfied: the class of states (e.g., attitudes) categorized as examples of social sustainability should be as homogenous as possible. This suggests that, other things being equal, we should prefer a theory that defines “social sustainability” in terms of one criterion only, or in terms of a conjunction of different criteria, to a theory that defines it in terms of a disjunction of different criteria. This also means that we have reason to aim for a monistic definition of “social sustainability” instead of a pluralistic one. The simplicity condition would thus not support complex definitions, which include some of the most common definitions, such as that of Barron and Gauntlett (2002).

3. Normative adequacy and value

Three conditions remain in our discussion of a good definition: the value condition, the condition of normative adequacy, and the amorality condition.¹⁷

The value condition is based on the idea that a good definition ought to help explain why we have value-based reasons to promote social sustainability and combat social unsustainability.¹⁸ For instance, we have reason to alleviate health disparities in a city because they, ceteris paribus, are unfair. Most participants in the debate on social sustainability believe that achieving it has positive value and that socially unsustainable communities are at risk (e.g., McKenzie 2004: 13; see also Vallance et al. 2011: 346, who support the value condition explicitly). In this sense, the value condition can be considered a special case of the ordinary language condition simply because, according to ordinary usage, social sustainability has positive value. To be clear, this implies that a definition of “social sustainability” that is not value laden – in accordance with Beckerman’s (1994) claim – fails to satisfy the value condition and is flawed in this respect.¹⁹
 

The evaluative judgments, with which the definition should be in line, are plausibly well-considered judgments, such that they reflect what we hold after some reflection. As with the ordinary language criterion above, the idea is that we should give considerable thought to what we believe is good and right. Then we can employ these conclusions to assess the definition of “social sustainability”. Using different types of examples (or even thought-experiments) can help us determine the definition’s implications.

The classic argument in favor of eating meat can help us understand why we should use well-considered judgments. Consider the argument that “it is permissible to eat meat because we have always done so.” After a bit of reflection, many of us would reject this argument. Throughout history, many horrors have resulted from what we “have always done,” such as subjugating women, slavery, and torture. A well-considered case in favor of eating meat thus cannot be based solely and without qualification on this argument.

In addition, there are value-laden and pragmatic reasons to use well-considered judgments when defining “social sustainability”.²⁰ In relation to value-laden reasons, employing well-considered judgments can be regarded as desirable because it means using justified judgments when reflecting on how we should structure our society. Well-considered judgments are often more justified and (perhaps) closer to the truth (e.g., McMahan 2013: 111–115, Tännsjö 2015: 1–18).

Promoting social sustainability measures is arguably easier from a pragmatic standpoint if the concept is based on well-considered judgments, because the reasons for using such a concept are stronger than for one based on immediate gut reactions. Policies that rely on a definition founded on well-considered judgments will possibly be both more stable and more acceptable over the long term. Moreover, the reasons for using such a definition will be much clearer to decision-makers. Of course, the potential problem with using well-considered, instead of spontaneous, judgments is that the definition of “social sustainability” moves away from the ordinary use of the term. However, this is probably a reasonable price to pay for the benefits – such as greater stability and higher value – related to using well-considered judgments.

The condition of normative adequacy stems from the idea that a definition of “social sustainability” should help us make better decisions. This requires the concept to work well when applied in practice. The results of applying the concept should be good and just, according to plausible normative views. For example, residents in a neighborhood identified as socially unsustainable might feel highly stigmatized. This might lead to less, rather than more, social sustainability being generated when applying a certain concept. The normative adequacy condition is therefore probably one of the most important criteria for devising a definition of “social sustainability” from a practical standpoint. Neither should the term be defined in such a way that it has the same extension as another concept, because that would make it redundant or ineffective. For instance, if our definition of “social sustainability” selects the same states as our definition of “justice,” then that definition of “social sustainability” does not help us make better decisions, since we already have another concept fulfilling the same function.

Finally, it seems reasonable that identifying an outcome as socially sustainable is not an all-things-considered moral reason for bringing about that outcome, but rather, if anything, a pro tanto moral reason. The definition of “social sustainability” should not conceptually presuppose that bringing about a socially sustainable outcome is always morally right. This final desideratum is often called the amorality condition. Of course, one could have an ethical view that it is always morally right to bring about socially sustainable outcomes, but that view seems highly questionable. (For example, imagine if the only way to bring about social sustainability was to sacrifice the lives of innocent people.) Moreover, the question of what role social sustainability plays in ethical theory should not be confused with (or built into) the conceptual question of how to define the term.²¹ In conclusion, the amorality condition requires that social sustainability does not entail moral rightness, and a definition that fulfills this can be consistent with different ethical principles describing the moral status of social sustainability.

The above considerations imply, first, that it is not always morally right to bring about socially sustainable outcomes, since there are other morally relevant factors than social sustainability. For instance, imagine someone with libertarian beliefs who thinks that outcome A would be more socially sustainable than outcome B, but achieving A would necessarily involve forcing people to pay more taxes. In the libertarian’s opinion, we should choose A only if we have everyone, or most people, on board with paying more taxes (i.e., we should not force people to pay taxes). Since enforcing increased taxes is morally unacceptable, the libertarian concludes that although A is desirable in terms of social sustainability, given the circumstances, we have an all-things-considered moral reason to opt for B. According to her, this is true at least until we have everyone, or most people, on board with the tax increases in A.

Second, a normatively adequate definition of “social sustainability” needs to accommodate the fact that social sustainability judgments on different levels of analysis do not need to converge. For example, it is sometimes argued that for a region to be socially sustainable, certain key services, such as primary schools, banks and pubs, need to be located close to residents (e.g., Dempsey 2009: 292f). Consider a place where very few people live and where maintaining these services is costly. We would see a strong tension between residents here and those living in other less resource-consuming areas. In this case, it is reasonable to say that we have a socially sustainable hamlet (because of the local facilities), but we have a socially unsustainable region (because of the tension between the different areas).

Third, it is desirable that a definition does not conflict with major international ambitions and treaties, such as those of the United Nations (UN), the European Union (EU), or similar large-scale governing bodies. A definition that conflicts with such ambitions and treaties is less likely to be used, and there is an increased risk that its positive effects will be nullified. For example, according to Murphy (2012: 19f), this means that we should include linkages to environmental sustainability in the definition of “social sustainability”. This is because it could shed valuable light on sustainability as such and it is sought after in key documents, such as the UN Commission for Sustainable Development (UNSD 1996) and EU Sustainable Development Indicators (Eurostat).  

To summarize, we have nine conditions of adequacy that form our theoretical framework for defining “social sustainability”: ordinary language, coherence, precision, reliability, measurability, simplicity, value, normative adequacy, and amorality. A definition that fails to satisfy these conditions warrants questioning, or revising, it, as shown in Figure 2 below.