This paper analyses the temporal unit of egalitarian concern. In the intertemporal context, the differences between egalitarian views can be appreciated not only in inequality analysis but also as regards the ideal egalitarian distribution to be established. In this paper, three intergenerational egalitarian principles (Complete Lives Egalitarianism, Corresponding Segments Egalitarianism and Simultaneous Segments Egalitarianism) are analysed and CSE is argued to be the appropriate egalitarian benchmark. The relations between the three principles and other moral ideals, namely maximin and utilitarianism, are also analysed. It is proved that CLE and CSE are compatible with a concern for the worst off and, partially, with a utilitarian concern, while the adoption of SSE implies a worse trade-off between egalitarianism and the other moral ideals.

Equality of when

What does it mean for a machine learning model to be ‘fair’, in terms which can be operationalised? Should fairness consist of ensuring everyone has an equal probability of obtaining some benefit, or should we aim instead to minimise the harms to the least advantaged? Can the relevant ideal be determined by reference to some alternative state of affairs in which a particular social pattern of discrimination does not exist? Various definitions proposed in recent literature make different assumptions about what terms like discrimination and fairness mean and how they can be defined in mathematical terms. Questions of discrimination, egalitarianism and justice are of significant interest to moral and political philosophers, who have expended significant efforts in formalising and defending these central concepts. It is therefore unsurprising that attempts to formalise ‘fairness’ in machine learning contain echoes of these old philosophical debates. This paper draws on existing work in moral and political philosophy in order to elucidate emerging debates about fair machine learning.

/pdf/fairness-in-machine-learning-lessons-from-political-2htlya0om1.pdf

Fairness in Machine Learning: Lessons from Political Philosophy

We review the theory of fairness as it pertains to concretely specified problems of resource allocations. We present punctual notions designed to evaluate how well individuals, or groups, are treated in relation to one another: no-envy, egalitarian-equivalence, individual and collective lower or upper bounds on welfare, notions of equal or equivalent opportunities, as well as various families extending these notions. We also introduce relational notions specifying how allocation rules should respond to changes in resources (resource monotonicity), technologies (technology monotonicity), preferences (welfare domination under preference replacement), and population (population monotonicity, consistency, converse consistency). We investigate the implications of these properties, in various combinations, in the context of various models: the “classical” problem of dividing an unproduced bundle, economies with production, economies with public goods, economies with single-peaked preferences, economies with indivisible goods, and economies in which the dividend is a non-homogeneous continuum. We consider economies in which resources are owned collectively, economies in which they are owned privately, and economies in which ownership is mixed. We offer a number of characterizations of particular allocation rules.

/pdf/chapter-twenty-one-fair-allocation-rules-5fdbxl4ikl.pdf

Chapter Twenty-One - Fair Allocation Rules

We study the difference between the ex post and the ex ante perspectives in equality of opportunity (EOp), and the possibility of a clash between them. We argue that ex ante EOp is a potential trap because someone motivated by ex post EOp may be led to believe that ex ante EOp is another natural embodiment of the same idea. As we show, it is not. Moreover, we explore the relationship between the ex post/ex ante tension and the well documented clash between the "compensation principle" and various "reward principles": we show that the tension between reward and compensation only exists if one endorses an ex post view of EOp; on the contrary, it vanishes if one adopts an ex ante view of equality of opportunity.

Ex ante versus ex post equality of opportunity

Resilience building has become a growing policy agenda, particularly for urban risk management. While much of the resilience agenda has been shaped by policies and discourses from the global North, its applicability for cities of the global South, particularly African cities, has not been sufficiently assessed. Focusing on rights of urban citizens as the object to be made resilient, rather than physical and ecological infrastructures, may help to address many of the root causes that characterize the unacceptable risks that urban residents face on a daily basis. Linked to this idea, we discuss four entry points for grounding a rights and justice orientation for urban resilience. First, notions of resilience must move away from narrow, financially oriented risk analyses. Second, opportunities must be created for “negotiated resilience”, to allow for attention to processes that support these goals, as well as for the integration of diverse interests. Third, achieving resilience in ways that do justice to the local realities of diverse urban contexts necessitates taking into account endogenous, locally situated processes, knowledges and norms. And finally, urban resilience needs to be placed within the context of global systems, providing an opportunity for African contributions to help reimagine the role that cities might play in these global financial, political and science processes.

Inserting rights and justice into urban resilience: a focus on everyday risk in cities in the South

In his Tanner Lecture of 1979 called “Equality of What?” Amartya Sen asked what metric egalitarians should use to establish the extent to which their ideal is realized in a given society. What aspect(s) of a person’s condition should count in a fundamental way for egalitarians, and not merely as cause of or evidence of or proxy for what they regard as fundamental? In this study I examine answers to that question, and discussions bearing on that question, in recent philosophical literature. I take for granted that there is something which justice requires people to have equal amounts of, not no matter what, but to whatever extent is allowed by values which compete with distributive equality; and I study what a number of authors who share that egalitarian view have said about the dimension(s) or respect(s) in which people should be made more equal, when the price in other values of moving toward greater equality is not intolerable. I also advance an answer of my own to Sen’s question. My answer is the product of an immanent critique of Ronald Dworkin, one, that is, which rejects Dworkin’s declared position because it is not congruent with its own underlying motivation. My response to Dworkin has been influenced by Richard Arneson’s work in advocacy of “equality of opportunity for welfare,” but my answer to Sen’s question is not that Arnesonian one, nor is my answer as well formulated as Arneson’s is.1 It needs much further refinement, but I nevertheless present it here, in a rough-and-ready form, because of its association with relatively fi nished criticisms of others which I think are telling. If this study contributes to understanding, it does so more because of those criticisms than because of the positive doctrine it affi rms.

On the Currency of Egalitarian Justice

I argue that desire satisfaction theories of welfare are not committed to the view that changes in welfare levels can happen after death, or that events that occur after death impact the agent's welfare levels now. My argument is that events that occur after death have only epistemological import. They may reveal that the person was successful (unsuccessful) in life, but the desire was already frustrated or satisfied before the person died. The virtue of the account is that it gives us a way to acknowledge both the intuition that we cannot be harmed after we die and, in a sense, the intuition that things that happen after we die are relevant to our lives.

Dead and Gone

Abstract COVID‐19 has significantly impacted expatriates, affecting their ability to work effectively and their personal and family lives. This study explores their experiences in different phases of the pandemic, and their perspectives on the future of international living in light of such global disruption. Involving over 600 expatriates in 48 countries, the research shows shifts in motivation and priorities as a result of the crisis and reveals how family (extended and immediate) topped expatriates' concerns. The study examines if and how support systems worked and looks at coping mechanisms, skills, traits, previous experience, and attitudes which participants found useful. A common theme was “falling between two stools”—where expatriates were ineligible for support in either their home country or in their host country. Others described being “pulled in different directions” or “torn by a dilemma” with pros and cons in either choice. The study contributes to the questions about the future of global work with results of the survey revealing changing norms for expatriation and global mobility raising the possibility of more nuanced approaches to global projects and less binary options for global professionals and their families.

https://onlinelibrary.wiley.com/doi/pdfdirect/10.1002/tie.22309

“Should I stay or should I go?”—Why the future of global work may be less binary: Lessons on approaches to global crises from the experiences of expatriates during the COVID‐19 pandemic

Joyce L. Jenkins

Papers

On the Currency of Egalitarian Justice

Dead and Gone

“Should I stay or should I go?”—Why the future of global work may be less binary: Lessons on approaches to global crises from the experiences of expatriates during the COVID‐19 pandemic