Eduard van Hengel

Sustainability as Intergenerational Environmental Justice: a discourse ethical approach

CONTENT
Abstract
Introduction
I. Discourse Ethics
II. The deontological distinction between moral and ethical questions
III. Sustainable development
IV. Future Generations
V. Sustainability and liveability
VI. Justice to future generations
VII. Strong sustainability
1. Biodiversity
2. Renewable Resources - functional sustainability
    a) immissions
    b) emissions
3. non-renewable resources : quasi-sustainability
4. Flow sources
5. Stronger than sustainable yield
VIII. Applications
1. Environmental Utilisation Space
2. Sustainability in environmental policy
3. Uncertainty: precautionary principle or risk analysis?
4. against the survival argument
References

Abstract

In this article Jürgen Habermas's and Karl-Otto Apel's discourse ethics is used in the first place to disembroll the notion of 'sustainable development' along with the deontological distinction between questions of justice and 'the good life' into an intergenerational moral norm of ecological sustainability and an intragenerational ethical development project. In this way sustainability can be operationalized as an ethically neutral set of ecological constraints whereas development should be considered a project of contemporaries, to secure economic prosperity, social equity and ecological quality within the afore mentioned constraints. The Brundtland Committee's notion of 'sustainability' is thereby conceptualised as intergenerational distributive justice with regard to the use of environmental resources.
The deontological distinction between 'sustainability' and 'development' gives rise to a differentiation of two types of ecological problems: problems of sustainability and problems of liveability, each with its own logic of argumentation, characterized respectively by a vocabulary of rights, obligations, norms and constraints on the one hand and of values, needs, preferences, trade-offs and compromises on the other.
In the second place from the more detailed procedural rules discourse ethics formulates for the justification and application of moral norms some suggestions can be derived for the operationalisation of sustainability norms that cannot de facto be put to the scrutiny of the majority of those involved: future generations.
This conceptualisation will prove not only theoretically fruitfull, in the sense of bringing practical intuitions and rational philosophical notions into a 'reflexive equilibrium', but also practically useful insofar as it suggests arguments for interventions in public debates about environmental policy, by partly justifying, partly criticizing prevailing approaches.

Introduction

"No entries found" was the result of a literature search in the Environmental Ethics bibliography for the concept 'discourse ethics'. This article undertakes to fill this regrettable gap; discourse ethics, as deployed by Karl-Otto Apel and Jürgen Habermas can provide - as we will show in the following - a fruitful conceptual framework for the interpretation of the notion 'sustainable development'. After a short introduction in the outlines of discourse ethics we will first set out the characteristic deontological distinction between questions of justice and of the good life, in Jürgen Habermas's version.
Next the central formula of the Brundtland Commission will be interpreted consequently as a definition of sustainability as intergenerational justice with regard to environmental resources. The rational reconstruction of the notion of 'sustainable development' according to conceptual distinctions provided by discourse ethics can - in our view - be brought into a satisfactory 'reflective equilibrium' with the prima facie plausible notion of strong but thin sustainability, as defended by ecological economists like Daly and Jacobs; sustainability can be operationalized in the concept of an 'environmental utilisation space' by Opschoor. All this probably to the surprise both of Jürgen Habermas who, in his scanty references1 to ecological problems, happens to consider them as ethical rather than as moral problems and of Herman Daly who himself, ironically, uses 'deontology' as an invective to disqualify an opponent2.

I. Discourse Ethics

Like all ethics in the Kantian tradition Jürgen Habermas's and Karl-Otto Apel's discourse ethics is - in Habermas's characterisation - deontological, cognitivist, universalistic and formalist3

It is deontological inasmuch it considers a narrow concept of morality, focussing on questions of justice and not - as classical moralities - on the 'good life' at large. It is about 'ought' claims, not about what is preferable to do.
It is cognitivist in that it considers normative rightness as a truthlike validity claim that can be rationally adjudicated; it transforms conflicts of interests into controversies about the validity of moral assertions that can be discursively redeemed.
It is universalist as it claims to formulate non-ethnocentric moral principles with a culture- and contextfree validity: unconditional categorical imperatives.
And it is (like the categorical imperative) formalist and anti-foundationalist insofar as it doesn't formulate any concrete, substantive, however general moral judgment but only a procedural principle to justify impartially the validity of such judgments.

But contrary to the individualist bias of most Kantian reasoning, discourse ethics is radically intersubjectivist. Instead of burdening (with Kant) the individual moral agent with the task to evaluate monologically whether s/he can consistently will his rule of action as a maxim for all, discourse ethics links normative validity with the outcomes of a real practical discussion about the appropriateness of a hypothetical norm between all those affected by that potential norm.
Discourse ethics is a two-level theory4, distinguishing between a first, theoretical-philosophical level and a second, practical level where participants in real world practices are involved in the solution of actual moral problems.
At the philosophical level no concrete moral contents are at stake; to the philosopher accrues merely the task to specify and justify prescriptive contents that can serve on the second level as regulative ideas for the participants in moral discourses. Discourse ethics as a philosophical ethics does this by explicating the assumptions of participants in concrete social practices. By this division of labour over the two levels philosophical theory avoids any paternalism over the autonomous discussions of enlightened moral agents; moral theory can only render explicit under what conditions participants themselves can find rational answers.

Discourse ethics arrives at its dialogical equivalent of Kant's monological categorical imperative in two steps. It first derives a general Discourse Principle <D> from universal pragmatics, which subsequently in moral discourse gets the form of a Universalisation Principle <U> for normative assertions. The discourse principle <D> lays down as a general criterion of validity "Only those norms of action are valid to which all possible affected persons could assent as participants in rational discourses"5. This non-specific validity concept is still equally applicable to questions of pragmatic truth, normative rightness and ethical-existential value and has subsequently to be specified in several different ways6. It is derived from formal-pragmatic presuppositions, 'what participants in a discourse must - necessarily but implicitly - suppose of themselves if they are to consider their consensus (should they reach one) rationally motivated'7. Argumentation is thereby seen as a dialogical procedure whose outcome rests on the powerless force of the better argument only, and whose results will uncoerced/freely be considered binding for the participants the more they are convinced that unequal power nor differences in social status or intellectual abilities did disrupt their discourse. These idealising presuppositions underlie according to Habermas and Apel any communicative lifeform.

In moral discourse where norms can be justified if and only if equal consideration is given to the interests of all those who are possibly involved8 the general discourse principle <D> takes the form of a universalisation principle <U>, the dialogical equivalent of Kants Categorical Imperative: 'a norm is valid when all concerned persons or parties can accept the consequences and side-effects of its general observance for the satisfaction of everyone's interest'9. This moral universalisation rule follows by material implication from <D> plus the understanding participants in the discourse must have about what it means to discuss whether a norm ought to be adopted10. As a criterion for normative validity in general <U> is still indifferent to the distinction between moral rightness and democratic legitimacy.11
As a general procedural rule for the impartial argumentative testing of norms <U> certainly incorporates a highly demanding standard of rationality, but one participants in real discourses must at least believe they have approximately been satisfied if they are to consider the norm emerging from their discourse as probably justified. The rules <D> and <U> are at the same time counterfactual (one proceeds as if they hold while one never can know for sure that they in fact have been fulfilled) and actually operative in the non-ideal real world: they constitute argumentation by their being presumed. These discourse-ethical idealisations, all competent participants in communicative practices have a practical prereflective grasp of, represent more than regulative ideas (Kant) since they effectively constitute real discourses ànd are at the same time something to strive for. (They are derived from speech-act theory and - because of their status called quasi-transcendental or transcendental-pragmatic presuppositions; they form the heart of Apel's and Habermas's design of a post-metaphysical ethics.)

It is at this point that differences between Apel and Habermas show up, differences that in our opinion are not of much relevance for our use of discourse ethics. In short, these differences pertain to the role of philosophy and the status of normative validity claims. Whereas both Habermas and Apel call their approach 'pragmatic' inasmuch they consider normative validity claims redeemable to presuppositions operative in everyday's communicative praxis of argumentation, Apels 'transcendental-pragmatism' makes the stronger claim that anybody acknowledging the validity of some normative prescription thereby obliges herself to act accordingly, on pain of committing a performative selfcontradiction. Habermas charges Apel on this point with philosophical foundationalism, i.e. trying to make people act morally through the use of arguments whereas in Habermas's more modest, formal-pragmatic view reconstructive ethics can only show 'the moral point of view' and its universality while the reasons 'why be moral', the moral engagement can only be found in socialisation and the existential belonging to some concrete communicatively structured community. For Apel this is an anti-foundationalist evasion, surrendering the categorical prescriptivity of moral norms to contingent empirical circumstances, a sociological reductionism, belittling philosophy as a mere fallible reconstructive science12.

Discourse ethics seems - as we will show underneath - a promising way to approach the concept of sustainable development. E.g. it self-evidently incorporates in moral deliberation the interests of future generations, as groups affected by the consequences of present day environmental policies13. Moreover from its procedural rules suggestions can be derived for practical ways to do justice to the interests of groups that cannot realiter participate in pertaining practical discussions. And it promises a way to overcome the individualist bias of most liberal theories, which is acute because in a finite world no right to material resources can be assigned to individuals, independent from the size of the generation to which they happen to belong14.

But first of all we should look into the deontological distinction between norms and values, a distinction which - as we will show - is extremely fruitfull to decompose the concept of sustainable development into a universal, context independent moral norm of sustainability and the concrete situated ethical objectives of 'development'.

II. The deontological distinction between moral and ethical questions

The centerpiece of the deontological phenomenology of morality is its distinction between questions of justice and questions of the good life, a distinction we will later on identify with questions of sustainability and of development, respectively. This basically Kantian intuition appears with different authors under different names, i.e. as the distinction between
-    moral vs ethical questions with Jürgen Habermas,
-    as justice vs the good with Brian Barry15 and John Rawls16,
-    as thin vs thick or minimum / maximum moralities with Walzer17,
-    as the right vs the good or as obligations vs valuations with Lutz Wingert18

From Jürgen Habermas's explications of this 'deontological intuition'19 we borrow the following 11 points.

(1) norms - values

Morality is, in its deontological meaning, about obligatory norms that inform decisions about what ought to be done, in any case; it imposes equal and exceptionless rules of action. Ethical values, on the other hand express what is preferable or desirable to pursue; they indicate goals for individual or collective action, derived from evaluative statements. They represent conceptions of the good. Whereas moral norms have the status of duties or obligations, securing the rights of others, ethical values represent stronger or weaker recommendations.

(2) conflict - cooperation

Moral norms and ethical values constitute, according to Habermas20 solutions for two distinctive problems of the intersubjective coordination of actions: the peaceful settlement of interpersonal or inter-group conflicts and the cooperative pursuit of consensual goals. Observing universally binding norms directs our rule-governed action, while ethical reflection on the other hand facilitates the choice and effective realisation of plans of action. Values, goals, needs and preferences have a teleological meaning, they control our purposive action. Morality and ethics serve the coordination of external, resp. internal relations of groups.
Historically deontological morality can be considered as resulting from the loss of community, caused by the disintegration of traditional culturally-homogeneous society by the modernisation process. It represents the unavoidable and universal device21 for coping peacefully with the irreducible plurality of potentially conflicting substantive conceptions of the good22 which characterizes modern pluralist societies where no comprehensive religious or philosophical system can any longer claim a privileged position in public life23. The deontological dichotomy of norms and values is therefore sociologically reproduced on every hierarchical level of society where the collective action of some in-group impinge on the interests of outsiders, the highest level being that of the unknown number of generations constituting mankind.

(3) rules - ideals

Norms are concrete and specific behavioral rules that stabilize mutual expectations in case of social conflicts between individuals or groups with competing interests. Values and preferences on the other hand are like regulative ideals: they can be approached gradually by being operationalized successively in a series of concrete goals or desirable objective situations without ever being realized in their full meaning. Moral norms specify what an individual or collective actor should do or refrain from. Obligations are act-specific24, they do not suggest who an actor should be or what he should become.

(4) criterion: impartiality vs authenticity

We evaluate moral obligations after the criterion of justice, they should be fair or impartial to anyone concerned, guaranteeing all parties equal opportunities to pursue their individual aims. Of values, goals and ideals on the other hand we discuss whether they articulate our most genuine concerns authentically or sincerely.

(5) agency: intersubjective - subjective

Moral norms have an irreducibly intersubjective character, the object of moral reasoning are always actions as elements of interactions25; they refer at least to two actors, behaviorally affecting each other. Values, interests or goals on the other hand can be an expression both of the lifeplan of some particular socialized individual or of the collective form-of-life of some historically situated group, structured by its history, tradition and worldview. Ethical values are inherently tied to some - individual of collective way-of-life; they remain ultimately 'subjective' and possibly individual, and have only local validity, bound up as they are to particular cultural traditions or existential decisions.

(6) prescription: thin - thick

Moral norms specify only what we must (not) do, in any case; they do not recommend what we should strive for. Norms do no more than set the outer limits to what is acceptable behavior26, they give no substantial answer to every question of practical action or public policy27. As decision rules norms of impartial justice are fraught with an unavoidable incompleteness28; the moral, norm conforming character of an action can only show itself as a marginal aspect of purposive behavior, motivated by some ethical, extra-moral intention. Therefore, respecting the obligations of a universal morality does in no way imply that anybody should behave impartially at any time or even restrict what s/he is doing for his relatives to what he is prepared to do for anybody else29; it presupposes that anybody is acting in his or her own interest but in doing so respects the marginal constraints that define the space, rightly left free for the selfish action of anybody else. For this reason deontological morality is sometimes characterized as a thin or minimal notion, compared with the thick nature of comprehensive or substantive conceptions of the good. 'Justice draws the limit, the good shows the point'30. Thin morality is restrictive, thick morality is proscriptive (Walzer). Under conditions of pluralism moral conceptions can no longer be general and comprehensive at the same time.

(7) authority: absolute - relative

Moral norms claim an absolute and unconditional authority, Kant called them 'categorical imperatives', they are binding not just for us but for all that act in the same situation, irrespective of their intentions. They are in this sense universal. Ethical or existential valuations however indicate preferences that are relative to some particular historically situated group or individual, with its own cultural tradition or identity; they state 'what is good for us' or for me.

(8) compliance: discrete - gradual / digital - analog

Moral norms claim a binary validity, they are valid or invalid31 and they can be complied with yes or no32; they cannot be fulfilled more or less and their observance cannot be subjected to any optimalisation criterion. As such they resemble assertory statements which can be either true or untrue. Values on the other hand appear in preferential orders, they are more or less attractive as one can subscribe to evaluative statements up to a certain point. Different preferences can compete for priority, they can be ranked flexibly and satisfied gradually.
As a consequence, norms should not contradict each other, they must stay in a coherent relationship, i.e. build a system, whereas values can compete with each other for priority.

(9) justification: rational - contextual

For their justification moral norms cannot be derived from any particular conception of the good. Moral norms are addressed to individuals or groups that do not share any substantive common understanding of the good; their universal validity only reflects the abstract communicative character of humanity's lifeform, expressed in the necessary presupposition of the possibility to settle peacefully conflicts of interest. In order to provide a just, fair33 or impartial34 basis of agreement for people with divergent conceptions of the good, moral norms should be ethically neutral; 'the moral point of view' should reflect an understanding of 'what is equally good for all concerned'.
So, while morality can only be founded on actor neutral reasons, ethical objectives are legitimated on actor relative reasons. Values, goals and preferences refer to particular traditions, they embody the situated and historically unfolding identity of an - individual or collective - actor. Values and preferences not only differ between various actors, they can also differ and contradict each other within one (collective) actor. In-group preferences can become a problem for us, as individual preferences can become a problem for me likewise. Solution of such problems require the affirmation of an identity or a new self-understanding; such solutions can be derived from a hermeneutical (re-)interpretation of one's life-history or of the constitutive tradition of a group. Ethical reflection takes place from an ego- or ethnocentric perspective (in a non-pejorative sense) whereas moral knowledge emerges from an ideally extended and generalised we-perspective.

Therefore only moral norms admit of rational adjudication across differing traditions, life histories and cultures. Ethical questions, by contrast do not admit of such impartial treatment because they refer to what, from a 1st person perspective is in the long run good for me or us, even if this would not be equally good for all.

(10) Priority of the right over the good

The mutual relation between moral obligations and individual or collective preferences is guided by the absolute or lexical priority of the requirements of justice over ethical recommendations pointing to the good life. In specific situations other people's rights are always 'trumps that override collective goals'35 and not just interests with a relatively higher priority. So, between moral norms and ethical goals there can be no bargaining nor compromising, nor can moral constraints be considered as being in conflict with any ethical objective. 'Die abstrakte Frage was im gleichmäßigen Interesse aller liegt, übersteigt die kontextgebundene ethische Frage was das beste für uns ist' (Hab..) Im Kollisionsfall können Rechte nur durch Rechte, Pflichte nur durch Pflichte übertrumpft werden.

As a consequence of the binary, resp. preferential validity of obligations and values, these behave differently when in a concrete situation two of them prescribe contradictory ways of action. When in an unanticipated situation two general and prima facie applicable norms collide, not their validity but their applicability will be doubted36; one will be considered appropriate or applicable and respected fully whereas the other has to be regarded inappropriate, if necessary through changes in its ceteris paribus clauses. When however two values or preferences speak to a situation, both can be appropriate at the same time, each of them being realised up to a certain point.

(11) motivating power: weak - strong37

While ethical goals and preferences have a strong intrinsic motivating power for the agency of the individual whose identity they express, moral obligations in the deontological tradition have, from the very beginning of the Kantian tradition been criticized for their impotence in motivating the subject when his rational duties conflict with his enlightened self-interest38. It is a common characteristic of cognitivist moralities that they decouple the validity judgment about a norm from its prevalence as de facto rule of action. For the single actor the condition of 'general observance' implied in the rationality criterion for moral norms can be considered as a resolutive condition; acting according to a rational norm, without this condition being met, is supererogatory for the individual subject. Good reasons for the rational acceptability of a norm are insufficient - weak - motifs for its factual observance. Therefore, no universalist morality can do without the motivational backing of accommodating social contexts that institutionalize abstract obligations in concrete life forms, with daily routines, socialisation processes, identities, internalized conscience functions and public law, with their sanctions of moral reproaches, bad conscience and punishments. Deontological morality therefore should not primarily be seen as promulgating rules for the single individual but as a rational intersubjective source for legitimate institutions, public law and legal rules39.

conclusion:

Deontological moral theory pays for its universalist validity, its immediacy, specificity and categorical priority by a narrow notion of morality that gives no meaning to life and is only weakly motivating. 'The moral point of view projects a sharp but narrow circle of light'40.

    A 'moral community' in Habermas's terms therefore should not be seen as a community of people, groups or institutions sharing some common values but only as being the potential sufferers of material consequences of each other actions ànd wanting to regulate their relation of mutual dependence in a peaceful way, that is: by observing norms that are in the common interest. One could summarize Habermas's distinction by saying: morality gets its authority bottom-up by generalizing the common negative whereas ethics comes top-down by interpreting and concretizing the common positive. Morality stems from a (dutch) lotsverbondenheid or (german) Schicksalsgemeinschaft, the finding oneself nilly-willy in each others sphere of influence, which is a far more objective condition than expressed in the english solidarity. Ethics on the other hand comes from a common identity, which is a shared subjective condition.

    Somewhere41 Habermas makes the rather obscure remark 'Moralische unterscheiden sich von ethische Urteilen nur nach dem Grad ihrer Kontextabhängigkeit', a remark we find repeated by Rehg 'Moral/ethical discourses are indeed different in kind, but their distinction in practice is often more a matter of degree'42. Such remarks in my opinion tend to undermine the rigid distinction discourse ethics elsewhere makes between moral and ethical judgments. One could however interpret this remark as a suggestion to hierarchicalise moral theory. In that case one could imagine that on any hierarchical level groups, institutions or regions with different internal ethical orientations at the same time acknowledge common moral duties, identifying them on a higher level of integration as fellow-sufferers of each others actions (not: as companions in some joint project) since sharing some common room for action; whereas the same elements could each act as level of moral integration, defining moral constraints for lower rank sub-groups with their own identities and projects. Hereafter we will not pursue this project to hierarchicalize moral theory but only suggest the possibility to integrate it with - and thereby improving upon - Nortons endeavour43 to incorporate hierarchy theory as an encompassing ecological theory44 into environmental ethics.

III. Sustainable development

Though much has been written in recent years about the concept of 'sustainable development'45, in our opinion there still has to be brought to daylight a core of rational meaning in this conceptual assembly; a meaning that we propose to uncover by ordering sustainable development's prima facie associations with environment, equity and future generations - given already with the seminal formulation of the Brundtland Commission (WCED) in 1987 - in a deontological, more precisely: a discourse-ethical framework. Instead of accepting the wildly divergent meanings given to SD we suggest to distinguish sharply between the two component notions sustainability and development. We thereby follow Paul Thompson's suggestion46 to treat sustainability as something more than 'a buzz-word'47 or 'cure-all' and develop it into a concept that promises to promote philosophical unity48. For 'development' on the other hand we are inclined to accept a wide array of politically and culturally inspired interpretations.49.

Notwithstanding the fact that WCED is certainly imbued with an ideology of economic growth as solution for both the poverty and the environmental problem, and consequently acted politically as a legitimation of industrialisation, modernisation and westernisation processes that hardly pay attention to its ecological consequences, it should nevertheless be praised for 'sustainable development' as a conceptual innovation with a more precise and radical intellectual content than most of its quotations - be they supportive or criticizing - recognize.
While surely many formulations in the Brundtland Report are of a nebulous nature (Parayil), at least its central definition admits of a close reading and more radical interpretation than most proponents have realised until now.

    "Sustainable development is a development that meets the needs of the present without compromising the ability of future generations to meet their own needs".

Though the concept of 'environment' is strikingly missing in this central conclusion of a United Nations Conference on Environment and Development (UNCED) it seems obvious that the environment is considered here as 'the ability of future generations to meet their own needs', an ability which can be impaired by 'development', understood as the meeting of the needs of the present; generations50 in this formulations are symmetrically considered as groups which supply in their own needs, while making use of resources, provided by the environment that they have to share. The environment acts as an objective interface between present and future generations, with the potential of transferring ecological effects of actions of earlier to later generations. Intuitively, the Brundtland formula seems to require justice or equality between generations in ecological opportunities, an intuition we will try to rationalize henceforth. Sustainability acts - according to this interpretation - as a qualifier of the environmental impacts of any generation's development.
As one of a host of critics of the Brundtland formula Beckerman considers its definition 'totally useless since 'needs' are a subjective concept'51; idem Pezzey52. In a similar vein the Dutch Scientific Council for Government Policy judges the formula insignificant because the needs of future generations cannot be known to us. These critics in our opinion miss the point; their remarks even can be used to expose more clearly the far more objective content of the formula: "Sustainable development is a development that meets the subjective needs of the present generation without compromising the ability of future generations to meet whatever needs, unknown to us, they may have." The formula clearly does not asks us to care for the needs of future generations nor to promote their wellbeing, as so many suggest53; future generations are evidently supposed to advance their own wellbeing. What is asked from us is only to preserve the natural conditions any generation needs for its self-reliance.
We therefore also reject Redclifts suggestion that the Brundtland SD-concept implies the essentialist assumption that needs can be divorced from the development process instead of being part of it; the Brundtland formula - quoted above - doesn't in fact show any such assumption; it merely considers development as a process of meeting needs, leaving fully open the possibility that actually these needs are historically determined during the socially organised process of their satisfaction. Indeed, more than that: with the words 'its own needs' it shows its cognizance that every generation can only know and meet its own needs.
The Brundtland formula is indeed about environment, equity and futurity54 but what is more: about a very well specified relation between these concerns. It seems correct to read the Brundtland formula as being concerned with the environment insofar as equity to future generations matters, while it deals with equity within our generation - particularly between the rich and the poor - in the concept of development.
The words 'without compromising' reinforce our deontological reading of Brundtland: they certainly do not suggest that the interests of present and future generations, or of environment and development should be treated on a par (as equals). No weighing or bargaining seems proper here55, these words rather specify an asymmetrical relation in that the environment as a collection of options for the satisfaction of future generations needs should be respected, as a negative norm: not indicating what goals or development strategies one should pursue but only what limits should be acknowledged in pursuing these ends. With the words 'without compromising' the Brundtland Commission in our view is ordering two questions of equity, fairness or justice in a hierarchical way: it suggests that intragenerational equity (synchronous justice or justice over space) as a moral duty for the members of nations of one particular generation can and should be pursued within limits posed by intergenerational equity with regard to natural resources (diachronous justice, over the long time). This rules out as a principle that the needs of the worlds poorest could be alleviated by shifting its costs to future generations.
In the Brundtland formula therefore sustainability and development maintain a well-specified relationship that is obscured in many secondary texts. 'Sustainable development' is not development that pays also due attention to the environment. Neither is sustainability the aim or goal of development, in quite the same way as safety is not the goal of our participation in road traffic56. The environment (in its potential to do irreparable damage to an endless number of far away generations) represents not just one of many rivalling concerns, competing for our attention with other legitimate concerns such as care for the poor, democracy, social justice, equality of chances, cultural and economic prosperity.
The only feature one could criticize the Brundtland formulation for is that its two opening words suggest to be followed by a substantial definition of 'development'; the formula however is - in our opinion - only a definition of the qualifier 'sustainability' as equity towards future generations with regard to natural resources.

So, in the Brundtland formula the two types of normative questions that are so sharply distinguished in a deontological perspective come together in an orderly way: development, the satisfaction of needs as the ethical goal of a generation, and the moral imperative of sustainability as environmental justice between generations.

Since the concept of development, especially as used in the present international community, comprises the question of economic and social justice between the rich and the poor nations, the Brundtland formula can be taken as ordering two types of social justice: diachronic or intergenerational justice between generations and intragenerational or synchronic justice within one generation, but in a clear lexical or hierarchical ordering, the first delimiting the space for the second.

In this interpretation of 'sustainability' as intergenerational justice in the use of natural resources, 'sustainability' has only a narrow, ecological meaning; it regulates but the environmental consequences of any generation's actions. While 'development' certainly requires a broad spectrum of social, economic, scientific and political initiatives, 'sustainability' is only about the environmental impacts of these actions. 'Sustainability' gives no substantial orientation to the proces of 'development' but only imposes a marginal constraint upon its ecological effects. In particular 'sustainability' should not be considered as the goal of 'development', as meant by the Brundtland report. Unfortunately, our common parlance is playing tricks on us here since, like a driver exceeding the speed limit in view of a policeman will call it his 'goal' to reduce his speed, a society exceeding the sustainability norm will call 'sustainability' its goal. Norm-conformity however cannot concur with real goals without abandoning the deontological distinction between norms and ends.

Even Lafferty who gives the Brundtland formula a more careful reading than most other authors fails to grasp its meaning on several points. This generations ecological performance should not be assessed with respect to the needs of future generations (Lafferty, 189) - which indeed are hard to know, qualitatively and quantitatively - but with respect to the natural potential we leave them to meet their own needs. And, unpacking the Brundtland concept into four principles, Lafferty separates caution with the disruption of natural processes from the future generations's possibility to attain similar standards of living, which leaves him criticizing Brundtland for not prioritizing between the requirements of equity and nature, which in our opinion is precisely what the Brundtland formula does very well.
    Welfare of future generation is beyond our control. You can only pass on fysical requirements. We have not to guarantee their welfare but their capacity to produce57.

IV. Future Generations

To get a more adequate idea about the timescales involved in the requirement to prevent damage to future generations we follow for a moment Norman Myers58 who explains how mass extinctions of species - like the one mankind is causing nowadays - will not be made good within less than five millions years and possibly several times longer59, since - judging by the bounce-back periods in the wake of comparable mass extinctions in the prehistoric past - that will be the minimum time required for evolution to generate a replacement stock of species matching yesterdays abundance and variety. Far from concluding that this demonstrates how in fact nature knows of no irreversible changes, all changes being reparable on their appropriate time and scale60, Myers continues supposing that in the interim period generations will succeed each other any 25 years; in that case no less than 200.000 generations, supposedly averaging a 2½ billion people each, will be obliged (through what we do or don't do in the coming decades) to live in a grossly depauperized biosphere: to a total of 500 trillion (500 mln x mln) humans. Two centuries of industrial revolution evidently have done more harm to the ecosystem than a number of ice-ages did.
This observation contrasts sharply with conclusions of a number of moderate and self-appointed realist intergenerational egalitarians that admonish us not to oversize the responsibility of mankind which could not possibly extend more than a small number of future generations: Rawls 'a dozen', Passmore six, Avner-de-Shalit 'about ten' because beyond that they consider the future as ostensibly unknowable. Says Avner: 'the strength of our obligations to the future 'fades away', in proportion to the degree of moral similarity' (p.25-31).
Myers observation wittnesses the existence of effects of human action that are both knowable and extending far more than a 1000 generations in the future. The diverging timescales of Myers and these social philosophers represent the categorically distinct timescales of natural and social evolution. Future generations are dependent on our legacy, both for their natural and cultural inheritance. Content and structure of the present generation's responsibility however differ widely in both compartments of our legacy, depending on the timescales of the transmission principles in both domains: natural laws and cultural tradition. The strong notion of sustainability is - in our conception - precisely about these very-long-term effects of human action on the natural environment. Whereas we (indeed) cannot be sure what will be the prevailing needs and values ten or more generations from now nor whether that generation will appreciate at all the natural or man-made treasures we could bestow upon them, we certainly cannot be sure that any of the 200.000 generations to come would not appreciate them61.
At least mediated through evolution, our ability to do well and increase biodiversity is by far smaller than our potential to harm future generations, as restricting the natural scope of choice for them is far more simple than extending it. Sustainability is about refraining from actions with practically irreversible effects.
Though of course the period of 5 mln years is not quite certain, al least its order of magnitude is, and couldn't be as well 37 of 83.000 years.
In the light of this there are good reasons to strongly disagree with all those who regard it an overestimation of our powers to know and to predict, to be concerned with effects of our actions on generations several hundred years from now.
Present humanity's technological potential to do practically irreversible damage to an unending number of future generations via the environment seems to be the conceptual core of sustainable development. While it lies well beyond our powers to contribute to the well-being of far away generations, it is within our power to damage them. Sustainability is the negative obligation to refrain from such actions, a negative obligation yet, that requires some 'positive' policies and actions. (fading cultural proximity with nearly constant biophysical proximity.)

Notwithstanding the absence of reciprocity between us and future generations, and the one-sidedness (uni-directionality) of nature as a transmission belt for the negative side effects of our production systems to the far future, it seems hard - with the deontological distinction in mind -  not to read the Brundtland formula as pointing to a norm of justice or equity between generations, constraining the ecological effects of the development projects of particular generations, i.e. of contemporaries.

V. Sustainability and liveability

While conceptualizing problems of sustainability as long term environmental problems, we should recognize that not all environmental disturbances have the property to harm future generations; there are environmental disturbances without irreversible effects, effects that can be made undone within a period of - say - one generation, such as stench, noise pollution, smog and other threats to public and occupational health, organic pollution of rivers and swimming water, damage to the beauty of landscapes, animal welfare etc. These are all environmental effects that become problematic for us when judged against the preferences and value systems of our own culture. Most of them belong to the 'first generation' environmental problems which became acute because their adverse effects could be experienced by the now living; their solution often succeeded thanks to the transferring of reparation costs to future generations62. We suggest to call this second category of environmental problems63 'problems of liveability' because they stem from a society's concern for the quality of its living space.
This means that we can have environmental problems in both categories of normativity: moral problems of 'sustainability' and ethical problems of 'liveability', each of these having its own normative logic, and with a lexical priority of requirements of sustainability over requirements of liveability64. While 'sustainability' acts as a constraint on the possibilities for 'development', requirements of liveability are an aspect, internal to the notion of development, with the implication that such requirements can enter into trade-offs with other component aspects of 'development' such as economic growth. As we will show later on the moral/ethical distinction in environmental affairs also coincides with the scientific, resp. arbitrary character of the norms involved and with the pertaining principles to handle uncertainty: precaution vs risk.
The distinction between problems of 'sustainability' and of 'liveability' suggests a solution for the recent sharp controversy over dutch environmental policy between two dutch policy advisory councils. While the Council for Research on Nature and Environment65(RMNO), under the presidency of Opschoor introduced with a number of reports on 'sustainable development' the concept of a measurable Environmental Utilisation Space in the dutch discussion, the Scientific Council for Government Policy (WRR)66 criticised this approach as a technocratic appropriation of democratic competence by science: WRR obviously treated all environmental problems as questions of liveability and therefore as suitable for compromising with social and economical interests, whereas RMNO rightly sketched the special character of sustainability problems without duly emphasizing that not all environmental problems are of this kind.

VI. Justice to future generations

As a procedural theory discourse ethics does not allow for the derivation of any concrete action norms. Nevertheless, it can in a reflective way contribute to the practical discussions about the acceptability of hypothetical norms. For instance - as we have shown - by sharpening our power to distinguish between ethical and moral discussions, and - as we will try to show in the following - by specifying in a general way the criteria that rational norms have to meet.
This is all the more relevant since the wide diversity of future generations that have to bear the consequences of our long-term environmental disturbances cannot de facto participate in any practical discussion about the general acceptability of hypothetical sustainability norms, nor can they launch any social struggle as slaves, women, workers and blacks did in the past, thereby boosting the public debate on questionable ideas of justice. The practical discussion about norms of sustainability can only be a discussion among the privileged who have effective power over the future environment, without the moral right to use it. What norms should we respect in order that future generations could say 'they reasonably considered our interests'? Without the sheer possibility of social protest from deprived parties rational procedures will have to bear the weight in finding such norms. The value free sustainability norms we are after could be worked out, we suggest, by alternatively expanding a normative and a cognitive bridgehead until they fit each other, that is: by specifying on the one hand the criteria an intergenerationally impartial norm should meet, and on the other hand look into the elementary structures of the objective, natural medium that relates our generation to those in the far future. In the absence of discussion partners, the best we can do in our discussions is to refer explicitly to the criteria that justifiable norms should meet.

So in search for sustainability norms we are looking for a set of constraints upon our environmental interventions that should be neutral towards a multitude of differring and possibly conflicting need interpretations; they should entitle no generation to a privileged position within natural history. In discussing hypothetical value free norms we should try our utmost to regard matters 'from the moral point of view', that is: taking the perspective of all possibly concerned parties (generations) and asking 'what rules could be in the equal interest of all'.
We thereby can presume that the wider the variety of interests that have to be taken into account, the higher the level of abstraction our prescriptions deserve and the more abstract the concepts will be for formulating the behavioral rules that can legitimately - but always fallibly - claim universal validity.
We especially should take notice of Habermas's point that practical discussions should include 'the mutual criticism of the appropriateness of the language in terms of which situations and needs are interpreted'67; though there is no room for such mutuality with future generations, we at least could be aware of the possible cultural prejudices, intrinsic to the language system or the basic concepts we use. This in the first place implies that the norms sought for should be formulated as far as possible in terms of objective bio-physical properties of nature to be maintained; they should therefore be formulated in a maximally object-sided way. In a second line we could accept norms that restrict the objective effects of certain types or scales of technological intervention in nature. Completely in contradiction with this conceptual neutrality are norms that are formulated in social-contextual terms like economic utility, social efforts or political feasibility68.  Our intuition suggests that scientific descriptive concepts come, for the time being, the closest to the ideal of an interculturally stable and value free conceptual framework69.

In particular we could expect already beforehand that this search for value free norms of sustainability is incompatible with a usual (but not the only) procedure for establishing environmental norms: with the help of linear (or monotonous) dose-effect relations. In this common decision making practice scientists provide policy with information about a continuously rising level of adverse effects (like headaches, sleeplessness, diseases etc) while emissions are increasing, whereas politicians draw somewhere the line between acceptable and unacceptable effects, on a scientifically arbitrary point. Such decisions reflect a valuing of effects or a reasonable compromise between parties involved, such as employers and workers, or airline companies and local residents. Since in the case of sustainability norms the majority of parties involved cannot raise there voices while the now living should methodically suspend their value judgments, we have to expect another norm finding reasoning.

With these considerations in mind, we claim that a certain strain in the recent sustainability debate that can loosely be grouped around the notion of 'strong' sustainability70 comes closest to the practical discourse in search of our ethically neutral sustainability norms. In the following we summarize some of the - self-evidently fallible - results of this debate with some of the arguments that prima facie illustrate its empirical and normative adequacy.

In the literature we have in mind 'sustainability' is usually presented as pertaining to five categories of resources that have to be shared among generations, each with its own mechanisms for the transfer of effects over generations and consequently each with its special problems to find value free sustainability norms. We discuss them successively.

VII. Strong sustainability

1. Biodiversity

With biodiversity, as the corollary of genetic information, species and habitat variety, the cognitive question from an environmental perspective is: what could possibly be the effects of our instrumental actions on these generations in the far future, be it in a positive or a negative way? As Myers's contribution makes clear the long term effect of our assault on biodiversity is due to the slowness of the evolutionary processes of speciation and diversification and to the incapability of mankind to control or accelerate these processes (up to now, one could add, in an optimistic mood). It is not a contingent fact of the last two centuries that human society has only destroyed biodiversity, it seems that until now and within the foreseeable future the sheer possibility to create new diversity - on the higher levels of species and ecosystems - does and will not exist at all. Mankind's interference with evolution can only be destructive, in the simple sense of: reducing diversity. This means that our factual relation with people in the far future forestalls the possibility of doing any good to them; we simply do not have the opportunity to influence biological evolution into different directions with the result that we are relieved from the task to evaluate different possible outcomes of our creative actions. This simplifies - we contend - the search for sustainability norms, from the cognitive perspective.
Here we are very much helped by the de facto impossibility of any constructive interference with evolution: since along the evolutionary way we cannot do any positive, we should, in search of a universal rule for environmental behaviour, generalize all that could count as a negative experience or an encroachment on interests. As the first law of thermodynamics states that matter and energy cannot be destroyed, the nearest abstract categories, possible to interfere with in a negative way seem to be structure, organisation, complexity or low entropy. The notion of biodiversity seems not to be far off here. The Dutch RMNO suggests as a biodiversity maxim that 'human induced extinction should not exceed the natural background extinction rate' or 'hands off from biodiversity', keep all options open for the future or 'restrict your interventions in nature to the reversible'.
This norm seems at the same time a good approximation of the general criterium that no generation should be entitled to a privileged position in natural history; and it seems to be a practicable maxim as well.
With regard to the latter point we hasten to emphasize that this general maxim is far from (as Habermas would say) supererogatory or asking the impossible71. In a Dutch discussion the argument was used72 that already 99% of the species are extinct, so why be so devoted to the remaining 1%? This is a gross, malicious and demagogic misrepresentation of the facts. It obscures that, while there is a natural species turnover of 1 - 10 species a year, the total number of species reached its all times high at the beginning of the Industrial Revolution while the antropogenic extinction rate since than increased to more than a thousandfold of the natural extinction rate (zie Arrow e.a.). Though the RMNO norm requires a drastic reduction of habitat destruction, especially in poorest countries (who - still - happen to live amidst the species-richest habitats), it does not in any way imply that all species should remain available wherever they may have existed at any time. The norm pertains to the irreversible assaults on 'global biodiversity', leaving open the possibility to reversibly reduce 'local biodiversity', i.e. as long as such reductions could be rolled back with the help of restorative ecological knowledge73.

Our suggestion runs against Habermas's remark (though cursory and without much argumentation) that man's responsibility for [plants and] 'the conservation of entire species' cannot be inferred from any duty of interaction or morally' (Erl.225)
Surely, not as a duty towards (members of) these species but why not as a duty towards future generations?

    The difference between a lexical and a preferential priority seems to answer the question why 'biodiversity should be a first principle, trumping other moral principles' or 'the preeminent value, subordinating other worthy objectives such as enhancing the life prospects of the worst-off people.'74

2. Renewable Resources - functional sustainability

a) immissions

We now first define a more restricted notion of sustainability, 'functional sustainability' before subsequently discussing its relation to the ultimate notion of sustainability as intergenerational environmental justice.
We encounter the most characteristic sustainability problem in the use of renewable resources like the catch of herrings and whales, the harvesting of trees and forests, the use of groundwater and soil fertility. This type of resources are exemplary for the still controversial notion of a - scientifically measurable - 'environmental utilisation space' or ecospace as introduced by Opschoor. Renewable resources can in principle be harvested indefinitely, that is: as long as the sun shines, its backing regeneration systems remain intact and the yearly withdrawals do not exceed the yearly growth of the stock or population. Exceeding this sustainable yield (SY) makes the total population shrink, its yearly growth will consequently decrease and so next year's sustainable yield will be lowered. Continuous transgressing of the SY will deplete the whole population until it will disappear totally when sunk beyond its 'biological minimum' for survival: a step in the direction of extinction of a species or a local irreversible erosion or desertification of soil, the event that should be avoided for reasons of intergenerational equity.
So for renewable resources, on any moment there is a sustainable yield with the population somewhere between the biological minimum necessary for its bare survival and the biological maximum, determined by the carrying capacity of the ecosystem on which the population is dependent. This sustainable yield reaches a maximum, Maximum Sustainable Yield (MSY), somewhere below the biological maximum of the population since at that point the carrying capacity of the ecosystem is already constraining the yearly regeneration. The concept (M)SY or - what we will call 'functional sustainability' - has interesting normative and empirical connotations.
If in any year SY is exceeded, next years SY is lower and should not be exceeded to prevent a further decrease of the population; to let the population return to its former and higher level even harvesting below this (already lowered) second year's SY will be necessary. So, there are already utilitarian arguments to take the MSY as a norm for the use of renewable resources, and keep a population at its MSY level75. Forestry learned this in the 18th, agriculture in the 19th and fishing did not learn it up to now76. Besides that the MSY is related to the intergenerational fairness norm, introduced above, since it is the amount that can indefinitely be yielded without threatening to cross the irreversibility threshold; surpassing the MSY is on itself, however, not yet an assault on the rights of future generations since it still can be reversed by underspending the SY for some time. So between MSY and the requirements of intergenerational justice there is conceptually no identity but an intrinsic relation and empirically a buffering zone.
With regard to the empirical content of MSY it should be underlined that MSY is an knowable objective property of a natural resource stock, more precisely: of its reaction upon human harvesting interventions. Unlike exceeding the speed limit, as a conventional or socially constructed norm, trespassing the MSY has not just moral or legal consequences but is has direct empirical effects. MSY is a threshold beyond which a new phenomenon sets in: the declining of a population. MSY represents a discontinuity in the yield-effect relation, a threshold whose location can in principle77 be ascertained with scientific means, be it in practice only after a temporary transgression of the limit in question78. The eventual normative - moral or utilitarian - decision to take MSY as an action norm therefore does only involve the assigning of normative authority to this natural demarcation but not the decision where to draw the line; fixing the line by agreement somewhere under or above MSY doesn't make any sense, at least not for nature. MSY's so represent a tertium between on the one hand the 'strong' laws of nature (like the laws of thermodynamics that 'forbid' a perpetuum mobile or relativity theory that rules out travelling quicker than the velocity of light) which maintain themselves, and on the other hand 'soft', conventional norms with regarding to the environment (e.g. for road safety or occupational health) which can only be maintained thanks to the moral or legal consequences of their violation. The lawlike character of a MSY does not preclude its short term excess; in the long run however exceeding the MSY undermines its own possibility. In the case of MSY's the relation between Is and Ought is more intimate than with standardisation elsewhere. Daly79 and RMNO80 accordingly suggest as an input norm of sustainability for renewable resources: 'harvesting should remain within the regenerative capacities of the resource'. Also Norgaard81.
This special factual character of the norms for maximum sustainable yields has in our opinion been responsible for the above mentioned controversy between dutch advisory councils. While criticizing the idea of scientifically decidable norms as naturalistic fallacy or technocratic arrogance, WRR obviously presupposes the traditional normalisation logic where science can only inform politics about continuously rising costs or risks (linear dose-effect relations) and politics has to draw a line somewhere between the acceptable and the unacceptable. SY's however discriminate between the possible and the impossible, on the long term. About sustainability norms there can be no bargaining, trade-offs, compromising or 'balancing between ecology and economy'.

b) emissions

Pollution of soil, air and (ground-)water with toxic substances or natural substances in unnatural concentrations or on unfamiliar locations is, as an addition to the environment comparable with the afore mentioned extraction of natural resources. Here also ecosystems contain buffers and a regenerative capacity that make discharges up to a certain point (quantity, time, concentration etc.) indefinitely sustainable. Here also the limit of sustainability is scientifically knowable by the empirical consequences of its excess. A characteristic difference however is that whereas nature itself stops the possibility of extraction of a resource, nature cannot stop anyone with the continued dumping of foreign substances, far beyond its absorptive capacities. Here RMNO and Daly formulate, rather obviously, as the output norm that assimilative capacities should not be exceeded and no accumulation of foreign substances should take place. The words 'rather obviously' expresses our judgment that, from the intergenerational moral point of view it seems hard to challenge the impartiality of such a general norm, given that the alternative is 'keep systems intact that can deliver services for millions of years or let them go down'. We anyway do not deal with these several intergenerationally relevant domains of action to suggest that from a discourse ethical perspective the normative discussion can be finished, but only to illustrate the cognitive traits of our relation with unborn generations in the far future where moral reflection is needed.

3. non-renewable resources : quasi-sustainability

In the use of finite stock resources like fossil fuels (coal, oil, gas) or minerals (copper, chlorine etc.) problems of environmental justice can arise both at the input and the output side. At which of both sides the setting of norms is the most pressing is a contingent question.
At the output side emissions of CO2 or the unavoidable leakage losses along with the recycling of minerals can threaten the absorptive capacities of regenerative systems. In this case norms will be settled in the same way as with emissions.
In search for norms at the input side however we meet a rather particular problem: equally sharing a finite stock of resources between an infinite number of generations would strictly spoken allow for no use whatsoever during one generation: an absurd conclusion82. This implies that for non-renewable resources only 'quasi-sustainability' is possible. Daly suggests that[.................................]

Another suggestion was done by the Dutch RMNO; they propose as a norm that 'the remaining, economically exploitable stock should be kept on a level sufficient for 50 years use, an ingenious device, requiring a yearly cut-back in resource expenditures of not more than 2% , with the paradoxical result that indefinitely there will be enough for 50 years83. This rule conforms to the definition that functionally sustainable is the behaviour that can, with regarding to its environmental consequences, be continued in perpetuity. Once expenditures have risen to the 2% level of stocks, the norm asks for a permanent but rather feasible curtailing operation, whereby the social benefits of the decreasing extractions can be maintained by rising technical efficiency of the resource use, and even the netto withdrawals must not be reduced when the economically exploitable stocks are increased. The logic behind this rule is that, while the stock cannot be the object of equitable sharing, the burdens of the unvoidable shrinking of the stock should be fairly distributed. To formulate this intergenerationally neutral norm, we need a culturally independent standard of what in any society whatsoever counts as an effort with regarding to environmental stock resources. This explains in our opinion why not the objective geological stocks are taken here as standard but the economically exploitable stock which evidently is dependent on economic parameters; since after all we do not intergenerationally compare stocks here but efforts, it contributes to the conceptual neutrality of the criterion to specify these in terms of an (economically impregnated) effort to reduce the use of an (economically impregnated) resource stock. Of course, this intuition needs a more careful foundation.

4. Flow sources

It is interesting just to mention finally the only type of resources we have to share with an unending number of future generations, without the necessity of fixing any sustainability norm: flow resources. Basically the radiation of the sun and the wind and water power derived from it are by nature distributed over the generations in a way we could hardly improve upon, neither could any generation illigitimately appropriate shares of the future.

5. Stronger than sustainable yield

As we stated already, we consider Maximum Sustainable Yield or 'functional sustainability' as a more restricted notion of sustainability because it can go along with serious destruction of biodiversity and species extinctions as a consequence of the scale or the unintended negative side-effects of the technology, used in harvesting the resources. Extended monocultures or the effects of tuna fishing on the dolphin population are cases in point. Also in wild nature highly stable ecosystems not necessarily have a high complexity or biodiversity. So agro-ecosystems that are 'indefinitely sustainable' with regard to their own resource needs are compatible with a very impoverished natural variety. While they - because aiming at an optimisation of biomass production - have intrinsically a lowering effect on natural variety, their pure scale could deprive large areas for the very long term of any other human use, in this way seriously diminishing the options of future generations. In the same way the negative side effects of technologies used could do irreparable damage to populations or ecosystems which happen to be unimportant for the reproduction of the primary system.

So sustainability in its emphatic meaning is about sustainable yield (its minimum meaning) plus the additional criterion of not irreversibly diminishing options for the future to use and shape the ecosystem in different ways. Sustainability as intergenerational ecological fairness therefore cannot be reduced to the functional reproduction of society (Skirbekk); it goes beyond functional imperatives. I also implies that unsustainable growth is not eo ipso self-defeating or suicidal (Parker)84, this is only true for 'sustainable yields'.
Illustrative is the discussion in forestry science about the (in that discipline long-standing) notion of sustainability where the usual meaning of 'non-diminishing returns for future generations of foresters' only recently becomes gradually re-interpreted via 'maintenance of a diversity of forest-functions' to 'keeping open other possibilities for the future'.
This consideration in our opinion labels biodiversity as the most encompassing sustainability criterion; indeed, as Norman Myers said already, 'a league of its own'85.

VIII. Applications

1. Environmental Utilisation Space

From the foregoing indications for the possibility of deriving a set of value free norms of sustainability we conclude that it makes sense to speak of this set of norms as an Environmental Utilisation Space (EUS) or Ecospace. The notion of 'space' aptly expresses that sustainability norms have the strong and uncompromising constraining character that moral norms deserve over concrete human projects. The suggestion that this space can be objectively measured by scientific means is both correct and misleading. It is correct inasmuch these norms can and have to be formulated in scientific language, being this - for the time being - the most general, precise and intersubjectively reliable descriptive language with regard to natural events, but this argument still holds true for conventional norms like speed limits or acceptable alcohol percentages. Above this however, sustainability norms claim objectivity inasmuch as they are based solely on empirical knowledge of natural facts and of value free normative reasoning, not being impregnated by any culturally local need interpretation. The misleading suggestion, on the other hand, that an EUS could be scientifically measured in the same way as the volume of my room or the surface of the earth should be rejected.
On the other hand, the undeniable normative content of the EUS should not entice policy makers to mistake the EUS for, or enlarge it into the contingent set of policy induced environmental entitlements. EUS as the corrollary of strong sustainability has a policy independent existence, shown dramatically in the fact that this space has been exceeded and successively reduced even before the concept existed.

2. Sustainability in environmental policy

The greater part of sustainability norms can only be formulated on a level higher than that where policy measures should be taken or implemented. E.g. as a strict sustainability norm biodiversity preservation does only hold with regard to global biodiversity; local biodiversity reduction need not contradict that norm though it could be opposed on liveability motives. Many sustainability prescriptions therefore will affect an action situation in the form of an obligation to international treaties or other higher level agreements. In many cases, i.e. when treaties have legislative authority, they thereby automatically get the lexically higher priority that requirements of justice deserve over particular local strivings.
In the present situation however, where global and local communities live far beyond the norms of sustainability, the return to sustainable ways of production and consumption itself gets in ordinary language the status of a 'goal', as something that is worth aspiring for. This misleading idiom should not fool us into abandoning the distinction between requirements of environmental justice and desires for environmental qualities. The wish, desire or need to normconforming behaviour of course never can be treated on a par with need interpretations sui generis. For a car driver exceeding the speed limit, reducing his speed can be temporarily called his 'goal', but not a goal that has to be balanced with his desires to tank, lunch, see his sister and be back in office on time.
This seems particularly befuddling many discussions about the legitimation of projects of landscape architecture, nature development or ecological restoration projects which characteristically 'aim' at both types of 'goals' at the same time: they want to make society more sustainable as well as more liveable. In discussing these aspirations however, we should stick to the understanding that sustainability prescriptions are always a sine qua non of concrete projects, whereas other aspects of nature conservation or restoration, undertaken for aesthetic, scientific, historic or recreative motivations can be compromised upon. Requirements of sustainability therefore can be considered as laying a bottom in environmental policy. The metaphor of the tenant is instructive here: while obliged to some general care for the structure of his dwelling, in order to pass it on undamaged to incoming tenants, he is free to paint, wallpaper and furnish it to his own discretion.

3. Uncertainty: precautionary principle or risk analysis?

The distinction between sustainability and liveability questions is extremely fruitfull in approaching the problem of scientific uncertainty. Above we now and then stated that effects or thresholds were 'in principle' knowable in a scientific way; the words 'in principle' conceal that in fact certain or even adequate scientific knowledge is often missing. With regard to the limits of sustainability however this uncertainty does not open the door for economic-strategic or risk-analytic approaches. We contend that the characteristic uncertainty approach to sustainability issues (and to questions of justice in general) is the precautionary principle: "where there are threats of serious or irreversible environmental damage, lack of full scientific certainty shall not be used as a reason for postponing cost-effective measures to prevent degradation."86 Precaution asks for non intervention in undisturbed ecological processes unless its harmlessnes is proven. The principle implies reversing the onus of proof.87
Risk analytic approaches are acceptable when those who profit from the risky behaviour are identical with those who may have to bear the uncertain costs of the action. Transgressing of sustainability thresholds however implies passing the costs of one's action onto unconcerned generations. It belongs to the discrete character of moral norms that they confront us not with a continuum of gradually less attractive options but with a threshold that delimitates our room for legitimate action from that of others. The categorical character of moral obligations prohibits entering options beyond the line in strategic calculations. Sustainability problems are no cost-benefit analyses nor can they be replaced by that. Unjustified outcomes remain unjustified, however smal the chance to it.
It seems adequate to consider risk analysis and the precautionary principle as the uncertainty logics of moral, resp. ethical problems or of sustainability, resp. liveability questions.

4. against the survival argument

Sustainability as Intergenerational Environmental Justice is, just as little as the ' onstant natural capital stock' criterion (regardless its weak or strong interpretations) certainly not about the 'pragmatic minimal biophysical requirements for humane survival' or the 'biophysical assets essential to humankind'88. It seems nearer to the truth - though less alarmist - to say that for any level of material throughput that physically can be sustained by a - however impoverished - nature, a size of the human species can be indicated which, with sufficient smart technology and efficient social organisation can live a good life within these natural limits, whereas for any arbitrary rich natural environment a population figure can be found that, due to inefficient social organisation and technology will not be able to maintain its numbers. Let alone that it goes against our logic of responsibility and blame to attribute an eventually declining number of humans - of which anyhow the least privileged will always show the highest death rates - to a shortage of natural resources, since the human species as such cannot uphold any number at all without a certain level of social organisation. Even in the extreme case of humanity going extinct within a very short period of say some months because of the incidence of some very aggressive and unknown virus, the virtual reflectioners on this disastrous event would evaluate it in terms of humanity's inability to timely prognosticate or prevent the spreading of the disease or to isolate groups from it or to find therapies; that is: they would blame social organisation.
Generally: however serious the questions we discuss, to suggest that human survival is at stake is always empty rhetorical exaggeration. We therefore agree with Apel's comment on Jonas's ethics of responsibility that 'survival' cannot serve as a quasi-ontological ethical goal but that the discussion should be framed in terms of justice for the now and than living89.

(to be continued)

References

1    Jürgen Habermas, Faktizität und Geltung (FuG), Suhrkamp, Frankfurt, 1992, p.204, Erläuterungen zur Diskursethik (Erl.), Suhrkamp, Frankfurt, 1991, 219.
2    in a comment on Mark Sagoff, Carrying capacity and ecological economics, BioScience 45(1995)9,610-624
3    Jürgen Habermas, Erl. p.11vv; see also: Matthias Kettner, Scientific knowledge,discourse ethics and consensus formation on public policy issues H.7 in ....?(mrt.'94)
4    (Erl. H.5).
5    Rehg, Fact and Norm (F&N), 62/3
6    F&N 108
7   
8   
9    or: 'a norm is valid only if all affected can accept the consequences and the side effects its general observance can be anticipated to have for the satisfaction of everyone's interests (and these consequences are preferred to those of known alternative possibilities', Habermas, Disc.Ethics 65, Moralbewußtsein. 75/6, Einbeziehung. 60
10  ontleend aan Harald Grimen, Consensus and normative validity, Inquiry, 40(1997)47-62
11   F&N 107, FuG 138
12     See: Habermas, Erl. p.184-197; K.O.- Apel, Normative Begründung der 'Kritischen Theorie', p.15-65 in: A. Honneth et al., Zwischenbetrachtungen im Prozess der Aufklärung, (Philosophical Interventions zu 60th birthday); Matthias Kettner, Karl-Otto Apel's Contribution to Critical Theory, in: D.Rasmussen, (Ed.), Handbook of Critical Theory, Blackwell, 1994
13    'Betroffen ' nenne ich jeden, der von den voraussichtlich eintretenden Folgen einer durch Normen geregelten allgemeinen Praxis in seinen Interessen berührt wird '(FuG 138).
-    "I include among 'those affected' (or involved) anyone whose interests are touched by the foreseeable consequences of a general practice regulated by the norms at issue' (F&N 138, naar Skirbekk). Zie ook Einbez, 42, 'unter Einschluß künftiger Generationen'.
-    The notion of  'all affected' in principle includes future generations, it does not differentiate between the now and the not yet living, for the notion of affectedness differences in time of existence count as little as differences in place: it is only the fact of bearing consequences of some action that renders some (potential future) person into someone whose interests should be evaluated in the legitimation of the action. (My formulation, following Kettner)
-    within discourse ethics with its universalistic claims, inbuilt non-preference over space nor time, time-bias, equality of human beings through space and time, justice between generations needs no special 'original' principle in human nature (Barry, 362-363, In Sikora òf Justice), reflecting the operative intuitions of international panels, intuitive rejection of generational egoism, temporal myopia,  (O'Neill) temporal parochialism
14    zie hierover onderstaande opmerkingen over 'group rights'.
15    Brian Barry, Justice as Impartiality, Oxford, Clarendon, 1995
16    Theory of Justice, Politial Liberalism
17    Walzer, Thick and Thin, Moral argument at Home and Abroad, Notre Dame Univ.Press, 1994, 108 p.
18    Lutz Wingert, Gemeinsinn und Moral, Suhrkamp, 1993, p.131-166, Unterschiede zwischen moralische und ethisch Fragen.
19    Habermas, Remarks on Rawls Polit. Liberal., J.of Phil XCII(1995)3,109-131; Erl. 100-118; FuG 197-207, Erl 105ff, 177ff, 168/169, Einbeziehung, 40-42, FuG (310), Rehg (F&N) 93-98
20    FuG 174
21    Brian Barry, Justice as Impartiality, Oxford, Clarendon, 1995
22    Barry ±127
24    {16} Matthias Kettner, Rational foundations of moral responsibility for nature, in: A.Oefsti (ed.), Ethics and Ecology
23    'free and equal citizens are deeply divided by conflicting and even incommensurable and incompatible yet reasonable comprehensive religious, philosophical, and moral doctrines', John Rawls, Political Liberalism (PL), 1993, New York, Columbia Univ. Press Paperback Edition 1996
25    Wingert, 154
26    principles of justice set limits to permissible ways of life; claims to pursue ends that transgress those limits have no weight. (Rawls PL 190)
27    Barry
28    Barry
29    Barry, voor § 19
30    Rawls, PL, 174.
31    Habermas, FuG (310)vv
32    Wingert, 156
33    Rawls, Justice as fairness
34    Barry, Justice as impartiality
35    Dworkin, Taking Rights seriously, 1977, they 'trump' claims of general welfare (Dworkin, Liberalism, 113-143 in St.Hampshire (ed.), Public and Private Morality
36    see Klaus Günther
37    a.o. Erl. 135
38    A motivational deficit that is made much of in concepts as social dilemma and social trap, in Prisoners Dilemma approaches and rational actor modelling.
39    See FuG
40    Habermas, Nachholende. Revolution 118 (?)
41    Erläut. 219
42    Rehg 92
43   Bryan G.Norton, Context and hierarchy in Aldo Leopold's theory of environmental management, Ecological Economics 2(1990): 119-127; id., Toward Unity among environmentalists, New York, Oxford Univ.Press, 1991; Norton, Bryan G., Ulanowicz, Robert, Scale and Biodiversity Policy: A Hierarchical Approach, Ambio 21(1992)244-49; Bryan G.Norton, Bruce Hannon, Environmental Values: a place-based Theory, Env.Ethics (??)(1997)227-245
44   Allen, T.F.H., Starr, Thomas B., Hierarchy: Perspectives for ecological complexity, Chicago, University of Chicago Press, 1982; O'Neill, Robert, et al., A hierarchical concept of Ecosystems, Princeton, Princeton University Press, 1986; Allen, T.F.H., Hoekstra, Thomas W., Toward a unified ecology, New York, Columbia Univ. Press, 1992 
45      -    uncritical accumulation of meanings (Riordan,192)
    -    cannot only be a mobilizing concept, without content
    -    CPB waarde Brundlandt meer wervend dan analytisch
    -    Sust. 'a motherhood and apple pie' concept, so abused as to become meaningless (O'Riordan, 1988,30)
    -    J.Pezzey, Sustainable Development Concepts: an Economic Analysis, World Bank Environment Paper no 2, Washington DC, 1992 {46}
    -    a rhetorical talisman
    -    SD: besides being usefull as a green gloss on industr. develpm. (Christoff)
46    against Skolimowski's 'Some admonish us that we use vague concepts as a vehicle leading to the amelioration of the environment and of human life. Yes, we do. It is better to muddle through to salvation than to go crisply to damnation.' Skolimowski, Henryk, In Defence of Sustainable Development, Env. Values, 4(1), 69-70, 
47    Thompson, 179
48    Thompson 12. Following Thompson we should continue our elaboration of the sust-concept by delineating as accurately as possible its descriptive and normative meanings. Also Harvey Brooks: For the concept of sust in the process of developm. to be operationally useful it must be more than just an expression of social values or political preferences disguised in scientific language. Ideally it should be defined so that one could specify a set of measurable criteria such that individuals and groups with widely differing values, political preferences or assumptions about human nature could agree whether the criteria are being met in a concrete development program (1992, 30 Sustainability and Technology', Science and sustainability, IIASA, Vienna) And also Norton (1992, ecosyst, health, 98) vraagt 'a set of principles, derivable from a core idea of sustainability, but sufficiently specific to provide significant guidance in day-to-day decisions and policy choices affecting the environment
49    In this 'veritable minefield of slippery terms and hidden value judgements' (Lélé/Norg.363) at least some of the mines can be rendered harmless by avoiding 'the quick-sand of values and world-views' (Lélé)
50    For the time being we can do with a definition of  'generations' as 'a set of people who are more or less the same age and who live at the same period of history, usually regarded as having a span of thirty years', Avner de-Shalit, 138
51    Wilfred Beckermann, 'Sustainable Development': Is it a useful concept. Env. Values, 3(1994)191-209
52    Wijst Brundtland-def af vanwege 'needs',: John Pezzey, Sustainability: An Interdisciplinary Guide, Env.Values, 1(1992)321-362
53    e.g. WRR, thereupon dismissing the whole formula, since these future needs are impossible to know now.
54    Hayward,97; also Andrew D.Basiago, Methods of defining 'sustainability' Sust.Dev. 3(1995)109-119. Concl. sust is no particular end-state but a research methodology. Its essential components are futurity, equity, environmentalism and biodiversity.
55    as implied in the dutch NMP where  sustainable development is considered seaking an optimum between economy and environment. ('zoeken van een optimum tussen natuur/milieu en economie)
56    Is 'sustainability' the goal of sustainable development (Alan Holland, Natural Capital, p.169-182 in: Attfield, Belsey (eds.), Philosophy and the natural environment, Cambridge UP, 1994 
57    Commentaar Daly, Env.Values 5(1995)49-55
58   the discoverer of  'the hamburger connection' (How Central America's forests become North America's hamburgers, Ambio 10(1981)3-8) in: Economics of the Environment: a seismic shift in thinking, Ecol.Econ.15(1995)125-128
59     since this estimate rests on the - unwarranted - presumption that nature's restorative capacities are the same as with earlier biodiversity breakdowns. But 'the sheer loss of species may well turn out to be less significant in the long term than the reduction of evolution's capacity to generate new species' at a rate sufficient to repair the damage of the human caused mass extinction within the period of 5 mln years, which happened to be the 'normal' bounce-back phase after the five prehistoric mass extinction episodes' because of the near elimination of the biomes, serving as 'evolutionary powerhouses' in the past: tropical forests, coral reefs, wetlands. Present biodiversity crash has its impovering impact, not just on total biodiversity but above that on restorative capacities of evolution. See Norman Myers, The biodiversity crisis and the future of evolution, The Environmentalist 16 (1996)37-47.
60    as criticasters of a strong sustainability concept like to have it, see
61    Howarth, Chain of Obligation, it is not so much what each generation leaves its successor, 'chain of obligations', but leaves to/steals from an immense number of generations, without any intermediate role of successors than in procreating. Nature provides the interface between us and generations in far future, no intermediaries (chain of love (Passmore) or obligations (Howarth ) necessary., no social relationship,
62    see exchange Myers / Vincent-Panayotou:
63    with the name coined by the Dutch Council for Social and Economic Policy
64    The two recente enlargements of the domain of ethics (according to Habermas somewhere (?), referring to Patzig), i.e. to animals and future generations, therefore are of a different kind: questions of justice and of the good respectively.
65    Raad voor Milieu- en Natuur Onderzoek
66    Wetenschappelijke Raad voor het Regeringsbeleid, in its report Sustainable Risks: a standing Affair (?) of 1994
67   117 comm. Rawls
68   Opschoor's favorite example to illustrate the concept of an Environmental Utilisation space, the catch of herrings, is therefore hiding a number of conceptual difficulties, since with herrings there is accidentally a fair identity between the effects on nature (extraction of a number of herrings) and the intention or utility of the technology: marketing a number of herrings. The damaging effects on the seabed populations of mussel fishing with a trawl net indicate the need for a maximally object sided norm formulation.
69    This highlights the importance of attempts to develop a language to discuss human interventionism in nature with concepts overcoming the dichotomy between ecology and economy. See e.g. Malte Faber, Reiner Manstetten, John L.R.Proops, On the conceptual foundations of ecological economics: a teleological approach Ecol. Econ 12(1995)41-54
70    Herman Daly, the dutch RMNO under Hans Opschoor, Michael Jacobs (see Env. Val, 1995/??, Ecological Economics etc.
71    Even Norman Myers moves - without saying - the argument too quickly into the realm of the politically feasible, writing "We cannot realistically seek to preserve every last species" and "To save all is impossible." It seems worthwhile to underline that these impossibilities only exist at the so-called realistic political level and not in a bio-physical sense; they, in our opinion, can in no way imply that the hereafter introduced norm of a biodepletion less than some species a year couldn't serve as a realistic but radical moral standard to measure the progress of realistic politics. See N. Myers, AMBIO 22(1993)2/3, 74-79, Biodiversity and the precautionary Principle. Also Botkin considers a 'return to precivilized extinction rates' as possible, see Botkin, Discordant Harmonies, Oxford UP, 1990, ong. p. 180-190
72    by WRR
73    Marlin L. Bowles, Christopher J. Whelan (Ed.), Restoration of endangered species; conceptual issues, planning and implementation, Cambridge UP, 1994
74   
75   M.B.Williams, Discounting vs Maximum Sustainable Yield, in: Robert Elliot (ed.), Environmental Ethics, Oxford Univ.Press 1995, p.21-29
76    This gave rise to a public discussion in the Netherlands why the government should continue to control the economic optimum, instead of surveiling the biological minimum, see Ad Corten, Volkskrant 1997
77    The words 'in principle' refer to the problems of scientific uncertainty which we wil deal with later on, but where also the distinction between problems of justice and problems of values will give rise to two different principles in the dealing with uncertainty: the precautionary principle and risk analysis, respectively.
78    Ludwig, Hilborn, Walters, Uncertainty, Resource Exploitation, and Conservation: Lessons from History, SCIENCE, 260(1993)17&36
79    Towards operational principles of sustainability, Ecological Economics 2(1990)1-6
80    Opschoor, Weterings etc
81    Norgaard, Development betrayed (p.212) bijv. stelt bepaling van draagkrachtgrenzen (door wetenschap) gelijk aan het bepalen van drempels voor gezondheidsschade 'as if the relationship between dosage and human health had a clear break point below which there was little or no effect. Plimsoll lines. Noemt dit 'mechanistic determinism', static limits
    acceptable levels of additives, pesticides and radiation.
    threshold and break points only as usefull rhetoric, against Daly
    all limits are social constructs
    'not scientifically deteRmined thresholds above which drastic events begin to occur, even though they are presented to the public in this manner'.'limits metaphors'.
82    Barry
83    The number of 50 is of course arbitrary; the argument goes for any number.
84    Impoverishing growth.  vgl Daly/Goodland on timber, sust. yield
85    N. Myers, AMBIO 22(1993)2/3, 74-79, Biodiversity and the precautionary Principle
86    UNCED, Rio, 1992
87   Beck, Gegengifte, 1988
88    Adam Przeworski, Sustainable Democracy, Cambridge UP, 1995, 141 p. Hdb90. John Buell. Tom DeLuca, Sustainable Democracy, Individuality and the politics of the Environment, SAGE, 1996, 154
89    Pearce, Barbier, Markandya, 1990, p.3