Monday 21 January 2008

Metaphysics for Objective Valuation

Ontology
I will start by setting out the philosophical foundation of my argument, specifically the ontology: that set of assumptions about the nature of existence upon which the argument depends. I assume that what appears to exist does so because it is an objective reality, though our experience of it provides us with a ‘model’ (mental representation) through which we understand and appreciate it. Thus our view of the environment is one of an internal model, built through experience, but we are contained within an objectively real environment that would exist, though differently in detail, even were we not to. A major movement in modern ontology asserts that objects never exist independently, but I refer to ‘entities’ in this paper with the simpler Aristotelian meaning of substances (see Smith, 2001), primarily for simplicity of argument. What this means in plain language is that I shall refer to entities with the meaning of discrete independent physical objects, like fish, electrons, trees and books and will not deal with more complicated arguments about, for example, inter-dependence or collective objects. Also for simplicity, I will avoid entanglement in arguments concerning quantum mechanics or chaos theory, though I recognise the growing importance these and similar ideas from theoretical physics are having on modern ontology.
I assume that entities are all made up of energy/matter organised into ‘forms’ in space/time by ‘information’. Here, information specifically means patterns of data that are capable of doing synergistic work. Matter and energy are joined by Einstein’s Special Theory of Relativity and are thought of as the raw materials for real entities. Space and time are united in Einstein’s General Relativity, and for the present purpose it is sufficient to regard them as providing a physical context for existence. Data and ‘information’, though, require careful explanation at this point and it is primarily to Thomas Sebok’s work that I turn for definitions. Sebeok is known as one of the world leaders of semiotics - the study of communication of information through signs and signals. Of particular relevance here is his success in resolving confusion between ‘information’ meaning entropy (as it is usually understood in information theory) and ‘information’ as “sensible signage” - the concept at the heart of his semiotics (Sebeok 2003).

“ Entropy is the process by which energy is devitalised of its capacity to perform work through successive utilisation. In popular parlance however, entropy is usually identified with disorder, or probability, or both. According to a mathematical interpretation of thermodynamics, utilisation of energy in a closed system results in entropy (devitalisation), and simultaneously [my emphasis] disorder of the system, both of which processes were thought to be the most probable directions in which nature moved. Hence entropy, disorder, and probability, were usually understood as synonyms for each other. Only with Ilya Prigogine's Nobel Prize winning research into thermodynamics, do we now understand the limits of earlier approaches to energetics. In real-life open systems, disequillibrium conditions can actually create order, which explains evolution, and disproves the inevitability of chaos. Entropy still occurs, because the expenditure of energy fuels the building of order, but the postulated equivalence of entropy, disorder, and probability is now an obsolete conceptualisation [Prigogine & Stengers, 1984].”

Further rejecting the definition of ‘information’ as the antithesis of entropy, on the grounds that the latter is best suited to strictly dealing with thermodynamics, Sebeok appeals for more precise terminology expressing the ideas of dissolution and integration in each of the three fundamental elements of existence (he calls ‘realms’), producing the following table:


MATTER ENERGY INFORMATION
anaboly essergy synergy
cataboly entropy dysergy
“Within each realm, the upper term in the column denotes integration, whereas the lower term denotes dissolution” (Sebeok 2003).

I follow Sebeok’s interpretation and use ‘data’ to represent an assembly of symbols without meaning and ‘information’ to represent data ordered so as to convey meaning - what Sebeok defined as “sensible signage”. In this context, ‘meaning’ is the property of information which enables it to perform synergy. The most relevant example of synergy here is the ability of information to act on matter/energy within space/time to produce ‘form’.

Indeed, ‘form’ is defined by Turchin (1999) as the ordering of mater and energy in space-time:
“The forms of matter we see in nature are subsets of the total set of all possible configurations of any physical system.” (Turchin, 1999).
According to Turchin’s systems philosophy, ‘metasystem transition theory’ explains how forms in general evolve through selection for a generalisation of biological fitness which roughly equates with stability. (Turchin, V. 1977).
“Relatively stable "systems" are constructed by such [evolutionary] processes through the mechanism of variation and selection. This leads to the spontaneous emergence of more complex organizations during evolution: from space-time and elementary particles, to atoms, molecules, crystals, DNA, cells, plants, animals, humans, and human society and culture” (Turchin 1991).

Smith’s “Elvis Presley” illustration explains why it is information, much more than matter that makes entities what they are. Referring to Aristotle’s eighth property of substance, Smith comments that: “substance, such as Elvis Presley, may continue to exist even though there is no physical part which survives identically from the beginning to the end of his existence.” (Smith 2001). Clearly we do not recognise the atoms making up Elvis, nor the space and time which he occupied, but rather the information which determined how these were ordered. An important corollary is that information alone creates and defines the difference among entities, as two x and y differ if and only if the information creating each is not identical - it is nothing to do with matter, energy, space or time. (In this sense matter and energy are what Aristotle called accidents: fat Elvis is the same Elvis as thin Elvis).

To summarise, we have real and objective entities composed of matter and energy, ordered in space and time by information so as to give them form. This information has the property of synergistic potential, enabling it to ‘make’ a stable form, when it is said, by the cybernetic philosophers (following Turchin), to be selected by its ‘fitness’ (which relates to the length of time a form can exist). Another point of view is that the information giving form to a stable entity is a ‘resonant’ subset of data: resonance being a subset of selection in the sense of cybernetic philosophy. Either way, it is the selection of a certain set of symbols (data) which when applied to the formless raw materials of physical existence, give us entities which we may or may not value.


References

Prigogine, Ilya & Isabelle Stengers “ORDER OUT OF CHAOS”. Bantam Books, New York, 1984.


T. Sebeok (2003) “WHAT I S INFORMATION ANYWAY?” in W. Sheridan SENSIBLE SIGNAGE 3rd Edition. URL: http://www3.sympatico.ca/cypher2/ChapterFour.htm

Smith B. (2001) Objects and Their Environments:
‘From Aristotle to Ecological Ontology’ in Frank A., Raper J. and Cheylan JP. (eds.), The Life and Motion of Socio-Economic Units (GISDATA 8), London: Taylor and Francis, 2001, 79–97.

V. Turchin (1999): "The physics of evolution", in: F. Heylighen, C. Joslyn and V. Turchin (editors): Principia Cybernetica Web (Principia Cybernetica, Brussels), URL: http://cleamc11.vub.ac.be/PHYSEVOL.html.

Turchin, V. (1977) The Phenomenon of Science" New York: Columbia University Press.

* Turchin, Valentin: (1991) "Cybernetics and Philosophy", in: Proc. 8th Int. Conf. of Cybernetics of Complex Systems, ed. F. Geyer, pp.61-74, Intersystems, Salinas CA.

Thursday 10 January 2008

On the existence and composition of intrinsic value.

In an earlier post (May 2007), I began developing a theory of information and value that seems to have universal meaning. Here I expand and rather more formalise the definitions of intrinsic value, information and the link between them. The purpose is to strengthen the theoretical foundation of my claim that information is the fundamental essence of all things in the Universe and that it seems to act with a will to live.

Introduction

Conventionally, value is thought to be an attribute attached to an entity by a person as an expression of their opinion about how much utility or happiness they expect to gain from it. This meaning of value is, by definition, subjective and in practice, ephemeral and subject to arbitrary change, making it an insubstantial and poor measure for public decision making. Any measure founded only on one or more people's opinion is not scientific because it is not objective and independently quantifiable. Conventionally defined value may provide a scientific measure of opinion among those making value-judgements, but cannot result in any scientific statement about the entities being valued, since it is devoid of objectivity. The work presented here is an attempt to find a basis for objective measures of value which are compatible with the conventions of scientific measurement. By implication, it is an exploration of the existence and composition of intrinsic value, since by definition, if an entity independently has an attribute which we may call value, then that attribute must be intrinsic to the entity. The familiar subjective meaning of value derives from an interaction between an observer (the valuer) and the entity under scrutiny; often it describes the observer more than the observed. An objective measure will depend strictly on the observed entity alone and can therefore be called its intrinsic value. We shall start by systematically defining various meanings of value.

Note: ‘Objective’ implies ‘A-Priori’ as used in the philosophical sense.

“I conceive that the great part of the miseries of mankind are brought upon them by false estimates they have made of the value of things”.
Benjamin Franklin 1706-1790

Different kinds of value.

“Nowadays people know the price of everything and the value of nothing.”
Oscar Wilde 1854-1900

1. Economic Value (Ve) is founded on the axioms of consumer theory:

A1.1 That consumers are 'sovereign', independent and free entities.
A1.2 That consumers always act to maximise their private utility and nothing else.
A1.3 That utility is gained from ‘consuming’ goods at rates which may differ among goods, but which are known to the consumer (also implying that the consumer is sensitive to change in utility).

The consequence of A1.2 and A1.3 is that goods can be substituted so that the maximisation of utility becomes the choice of a set (mix) of goods in proportions that yield the highest possible utility.

The consequence of A1.1 is that Ve is subjective in the sense that it arises from the properties of both the good being valued and the valuer.

Proposition 1: The result of this subjectivity is that Ve does not exist if there is no agent making a valuation.

Environmental economics defines value as the sum of use value and non-use value. The former consists of the utility derived from direct consumption and the latter is made up of passive-consumption (e.g. enjoying scenery or the pleasure of knowing the good exists) and deferred use value, including bequest value.

A consequence of Proposition 1 is that intrinsic value does not exist.

2. Innate Value (Vi).

“The things which are seen are temporal; but the things which are not seen are eternal.”
II Corinthians 4.1:18

To define Innate value, we must first define what I call Quantitative Existence (QE). Normally existence is thought of as binary (exist or not), but QE specifies the amount of existence in terms of time (how long for) and numbers (how many replicates of an entity exist). In biology QE closely approximates to reproductive fitness (lifetime reproductive success) and in Economics it may be equivalent to utility, but only for rational motivations (e.g. excluding harmful addictive behaviours).

Innate Value is now derived from the following axioms:

A2.1 Entities always act at time t so as to maximise their QE.

It is interesting to note that this matches the axiom behind the Selfish Gene Hypothesis and if genes are regarded as packets of information, then it amounts to a generalisation of that hypothesis in which information universally seeks to maximise its QE whilst the universal increase of entropy in time (2nd law of thermodynamics) is constantly diminishing the aggregate QE of information.

A2.2 For any entity x, the existence of at least one other entity y at time t enables an increase in QE(x).

Axioms A2.1 and A2.2 allow us to deduce that any entity x may attach a value Vi(t,x,y) to every other entity y in proportion to the contribution y contributes to the QE of x at time t.

Note that this definition makes no requirement for choice, or even detection of the influence of a good on QE, though it permits both.

Since choice is not necessary, sentience (in any form) is not relevant to Vi, which therefore may apply to any entity, including inanimate objects, and even non-physical entities, for example ideas.

This leads to proposition 2: All things value at least one other thing.

A hierarchy (or chain) of valuations can be conceived, leading back to the basis of all physical existence - the founding laws of the Universe (which are the laws of physics as far as we know).


3. General and specific innate value
We can divide innate value into two components which I shall call the general innate value Vig and the specific innate value Vis, respectively. Vig is the innate value of any entity with the properties needed to increase QE, for example plant food for a herbivore, whilst Vis is the value of an instance of this class of entities, for example a specific plant. More formally, Vig(x,y,t) ≥0 iff [ QE(x) | Y(t)≠0 ≥ QE(x) | Y(t)=0 AND y e Y]. In words this says positive Vig is assigned by any entity x at time t, if and only if there exists at time t, a member of a set Y of entities defined such that the existence of any member of this set y at time t leads to an increase in QE for x.

All entities in the class Y share innate value Vig and by definition may be substituted, one for another, since they all provide the same general attributes. Any additional innate value arising from specific properties of any instance of the class are unique to that instance and therefore contributes to Vis. Thus x will attach further Vis to any y having unique attributes that increase QE(x).

The composition of entities and the supremacy of information

Axiom : All things which exist are, in various proportions, composed of a) energy/matter, b) space/time and c) organised information.

Energy and matter are treated together because of their equivalence established through Einstein’s Special Relativity. Similarly space and time are treated together because of their relationship established by General Relativity. Information theory enables us to account for all bits of information as data, irrespective of whether it is meaningless noise (disorganised information) or, potentially meaningful sequence, often referred to as signal, that is organised information.

All things are what they are because and only because of the organisation of information. The forms and mix of kinds of energy/matter and space / time occupied are set by organised information and observing the form of the entity is equivalent to reading this information. In other words energy/matter and space/time are formless raw materials that are structured by organised information into entities.

From this we can deduce that class of entities Y all must share in common energy/matter and space/time, though perhaps in quantitatively different amounts. The class Y is defined by the organised information which structures these commodities and gives them form. Further, what makes instances y of Y unique is the additional information which their composition carries. Therefore Vis results strictly from organised information content. This conclusion can be generalised to Vig, by realising that a class of entities Y is simply a generalisation of the description of an entity by attributes. Information defined what was unique about y, giving it Vis, but information also defines what is different between class Y1 and class Y2 of entities, equivalently giving each a class-level Vis. Since classes can be nested in an hierarchy which eventually includes (in the broadest class) all entities, there is no intrinsic difference between Vis and Vig.

We can now say that it is the information which defines entities, gives them uniqueness and therefore which gives them innate value.

In general, both energy and matter are interchangeable and may take many forms; space and time the same (in the sense that different parts of space and time may be occupied), but each structure of organised information uniquely defines some aspect of existence and is therefore uniquely necessary and so not substitutable.

Proposition 3: Energy/matter and space/time are abundant and within their category are substitutable, but organised information is uniquely necessary for the existence of all things and is therefore not substitutable.

We now have that unique organisation of information gives rise to innate value and that it is uniquely attached to a specific entitity. This means that unique organisation of information has objective value because it is a property of the entity itself. Furthermore, this property is potentially quantifiable, leading directly to a basis for quantitative intrinsic valuation through accounting for organised information.

Proposition 4: the value of a thing can be objectively quantified by measuring the amount of unique organisation of information needed for its existence.

This leads to a definition of Organised Information Value (Vo), which is objective. Vo is proportional to the quantity of organised information contained by an entity. This remains potentially distinct from innate value and intrinsic value, but is a good substitute for both.

Note that since innate value is derived from unique information, when a package of organised information is replicated, the Vo is shared among its instances.


“What we obtain too cheap, we esteem too lightly; it is dearness only that gives everything its value.”
Thomas Paine 1737–1809

Objective value as a link between two entities.

Objectivity is not exclusive to intrinsic value; it can also measure the innate value of one entity for another. This may be objective if and only if the quantification of value is based on a measurement of the intrinsic properties of the two entities (that is, distinct from the subjective - opinion based valuation Ve). Objective value which measures the effect of one entity y on the QE of another x, I shall call the Dependent Value Vd(x,y) to distinguish it from intrinsic value.


Effective Value

For many goods, it is easy to show that Vd is infinite because dependence on them for existence is absolute. This result is of little use for practical calculations of value.

This inspires the use of effective as opposed to absolute dependent value. Effective value Vf has the meaning of how worried we should be at the loss of a good. I shall show how it gives an intuitive justification for using the concept of substitution in calculating value.

We need not worry about the fundamental laws of physics because we have no ability to threaten their existence, nor exclude ourselves from them. We can therefore define their Vf as infinity times zero, which is indeterminate - a more realistic, but perhaps no more useful conclusion. However, this calculation illustrates the principle to be used in deriving effective value. Let Vi(x,y) be the innate value of x for y and T(x,y) be the capacity of y to remove (beyond use) x, then the effective value of x to y is Vi(x,y).T(x,y).

Since all entities are assumed to maximise their QE (Axiom 2.1), none would remove a good x with positive Vi(x,y), unless this was necessary in order to substitute it with some other good z with Vi(z,y)>Vi(x,y). The extent to which a good is removed is therefore the extent to which it is substituted, which can be found simply from the ‘rate of substitution’ - a measure familiar from economics. Indeed, in consumer theory, the rate of substitution of money p for a good x is the price of x.

Effective Value of y for x is therefore substitutability of y by x multiplied by Vd(x,y).

Significantly, when this principle is applied to the intrinsic organised information value Vo, it is simple to see that the effective value is equal to Vo divided by the number of accessible instances of the organisation of information x. This leads to:

Proposition 5: Objective Effective Value of x may be measured by the amount of unique organisation of information needed for its existence, divided by the number of instances of x available.

Note, availability may be a strong limit on the divisor if location, or other limits to access are accounted for. For independently reproducing instances (e.g. biological entities, the relationship between numbers and reproductive potential (population dynamics) must also enter the calculation - see future value below).


“Earth provides enough to satisfy every man's need, but not every man's greed.”
Mahatma Gandhi 1869-1948

Collective Value: Relaxing the independence assumption

So far, definitions have assumed that entities are independent of one-another in their actions and assessments and have assumed nothing about the effect of a behaviour of one on another.
Innate value does not allow for collective concerns, neither does the classical definition of economic value. However, the doctrine of ecology is that all living things are interconnected in mutual dependencies and human beings are a part of this ‘web’. Acting as free agents, we can step beyond the proscriptions of classical economics and act in the common interest, substituting some of our private gain for the common good.

We may ask if this has an analogue in the more fundamental realm of universal values (Vi and Vo). If all the QE of all things depends on all other things, then the common good is always served by individual QE maximisation. However this is not generally true (e.g. predation, competition and parasitism). Since nature does not show significant mutualism of the kind advanced by human moral philosophy, we cannot take this as a fundamental rule in valuation (indeed it seems to be a special trait of social animals and humans in particular). Since, then, there is no such thing as universal collective value, we need to distinguish between scientifically determined objective value and socially agreed value, which remains subjective. It may be that we decide on culturally determined values which differ from the scientifically derived objective values - that is a matter for normative studies, not science.

Future Value

Removing a good at time t may affect QE for all time including and beyond t. This, though, is contained within a complete definition of QE, since it is defined for a finite-life non-reproducing entity over the its entire life, for a reproducing entity by the reproduction success of its grandchildren, but it is ill-defined for an infinite entity. Thus no additional calculation is needed for future effects as long as complete QE is used. The economic practice of 'discounting the future' is predicated upon the assumption that future substitution will be more efficient (provide higher QE) than conservation. This assumption is, in principle, testable for objective measures of value.

The pursuit of sustainability

The Search for Sustainability

There are famously as many definitions as there are people discussing this concept. I will try my best to be precise and practical.

First, sustainability in general means maintaining equilibrium, or more broadly maintaining a system within elastic limits, so that its characteristics ‘orbit’ about one or more ‘attractors’ in ‘state-space’.

Next, what is being sustained? Options include ecosystems, biodiversity, our ‘way of life’, or our ‘standard of living’. None of these are precisely defined at present. I note that the most important meaning of sustainability attempts to address ecological and socio-economic systems simultaneously, in the context of possible conflict between them.

Borrowing from Economics, I use the axiom that utility is the property that best summarises our standard of living and more generally, quality of life, and invoke the concept of substitutability from classical economic theory. I add the further axiom that utility aggregated over the human population positively relates to ecological ‘health’ (referring, for example, to our dependence on ecological services such as provision of atmospheric oxygen, clean water and food), since it is well known that these services decline with decreasing ecosystem health. Then we can seek to sustain aggregate utility, which implies maintaining ecological health, or finding long term economically efficient substitutions for the loss of this health.

Short term substitutions for small and local losses of ecosystem health are abundant, but long term large scale solutions to major losses are unknown. Therefore sustainability is the requirement to maintain aggregate utility, in part by maintaining the health - form and function - of ecological communities globally. Ecological science provides some evidence that this includes maintaining biodiversity.

To obtain a proper definition of utility, we would have to venture into political theory, since different doctrines emphasise their own contribution to human welfare: capitalism counts property, measured by money, socialism emphasises egalitarian virtues of self-worth and green political doctrine emphasises our relationship with the natural world. The truth seems to be a combination of all these is needed to maximise aggregate utility, but the mix is a matter of individual taste.

Policy makers might best be advised to ensure adequate opportunities for individuals to enjoy all three categories and especially to ensure that none of these elements is seriously compromised by human activity, by which I mean that their sources must be preserved. In specific relation to ecosystems, this obviously means maintaining them within elastic limits by controlling direct exploitation and the effect of ‘externalities’.

Capitalism counts utility in terms of property, equivalent to wealth, measured in money. Erich Fromm* notes that we now tend to define ourselves by what we have rather than what we are. Thus our self-worth, or self-valuation appears to be rooted in money value rather than the intrinsic value of ourselves. More generally, it is hard for the modern western mind to think of intrinsic value, but easy to think of market-exchange value, as Oscar Wilde quipped “Nowadays people know the price of everything and the value of nothing.” Suppose instead, we took seriously the notion of objective intrinsic value, based on unique organised information content (see earlier post on information and value). Then when asked how much a person was worth, we would readily answer that it was an enormous, practically uncountable amount and that the difference among individuals, although real was so small in comparison with the totals, that it was effectively unmeasurable.

If our way of life is harming ecosystems to the point that diminishes their long-term ability to provide necessary services, then we need to make substitutions - replacing harming behaviours for harmless means of gaining utility. Since the harming behaviours are those of exploitation and consumption of nature in the pursuit of utility from manufactured goods, we clearly need either to produce the goods from lighter resource use, or replace a proportion of capitalistic utility with one of the other two kinds. Social psychologists such as Fromm, go further to say that capitalistic utility is a con - it does not bring happiness, but has the character of an addiction and is alienating and destructive of human self-worth. In this case, it is evident that a substitution with socialistic or green utility will be beneficial, one might even say that for every unit of material wealth exchanged for a unit of social or green wealth, we gain in utility. Therefore there is a theoretical underpinning to the notion that sustainability is best sought by changing not merely our lifestyles, but our understanding of what makes us happy by changing the way we value things.

* ‘To have or to be’ Erich Fromm (1976). Abacus. ISBN 0 349 11343 2