Analysis-Synthesis

I. Analysis—synthesis
Roberto Poli
Abstract: The multi-layered complexity of the connections linking
together analysis and synthesis is unfolded against the theories of wholes
and their parts and levels of reality. The paper claims that entities con-
nected by a relation of existential dependence and not related by a part-
whole connection are linked by a whole-whole relation. In exploiting
some of Herman Dooyeweerd’s ideas, one type of whole-whole connec-
tion is discussed.
Key words: Ontology, analysis, synthesis, whole, level, windowing,
enkapsis, encapsulation, identity
1. Introduction. The Need for a New Vision1
To date, modern science has relied on an essentially analytic strate-
gy. Different sciences have been developed in order to efficaciously seg-
ment the whole of reality into classes of more or less uniformly connec-
ted phenomena. The guiding idea has been that phenomena occurring
within each class are more causally homogeneous than phenomena per-
taining to other classes, so that the task of explaining their behaviour
should be more easily accomplished. This divide et impera (divide and
rule) strategy has proved immensely successful, at least for some regions
of reality. Other regions have proved more refractory, for a number of
serious reasons. The first is that different regions may require different
types of causation, some of which are still unknown, or only partially
known. A second reason is that for some regions of reality the analytic
strategy of breaking items down into pieces does not work properly. A
third and somewhat connected reason is the lack of a synthetic metho-
dology.
The complexity of reality requires the analytic strategy of segmenta-
tion into categorically homogeneous regions. This first move is not que-
1 A simplified version of this paper has been published by Global Spiral, 10(5),
2009 (http://www.metanexus.net/Magazine/).

20
stioned. However, some regions contain only items that can be further
analytically segmented into pieces. These items are entirely governed by
their parts (from below, so to speak). Other regions contain items with
different patterns: they depend not only on their parts, but also on the
whole that results from them, and eventually also on higher-order who-
les of which they are parts (e.g., organisms, communities).
Mainstream science, however, has defended and still defends the
idea that sooner or later all types of reality will be analytically unders-
tood. The outstanding successes achieved by analytic methods, however,
come at a price: namely the fragmentation of all types of entities into
their lower-order parts, especially material parts. Not only are living en-
tities seen as nothing other than biochemical reactions, but also mental
phenomena are supposedly completely understood in terms of brain dy-
namics. Moreover, persons are seen as agents, and furthermore as indivi-
dual (i.e., isolated), purely egoistic agents, as if they were atoms within a
container.
More profoundly, the analytic process of fragmentation creates ap-
parently insurmountable barriers (1) between physical and biological
sciences, (2) among the natural, cognitive and social sciences, and (3)
among science, philosophy and religion. The world itself becomes frag-
mented and loses its integrity. The fact is that analysis may inadvertently
destroy the relational linkages that are crucial in the study of ‘living’ sy-
stems. Synthesis, on the other hand, is a natural procedure with which to
study emergence: the (unanticipated) relational connections that appear
when a multitude of component systems interact. To give but a single
example, it is enlightening to observe our wealthy societies: apparently,
the systematic adoption of purely analytic methods results in our wealthy
societies become wealthier, while people and societies themselves seem
to become more and more internally fragile.
The main problem is that at least some items cannot be fragmented
without losing relevant information. Organisms, persons, communities
are some of the striking cases. When they are fragmented and reduced to
their ‘matter’ they lose their most essential property: life, for organisms
and spirit, for persons and communities. Analytic methods deprive per-
sons and communities of their spirit.

21
Admittedly, our understanding of non-fragmentable items is still de-
ficient. Otherwise stated, there is no denying that a properly developed
method of synthesis needs to be developed. Many scholars doubt that it
will ever be developed or acquire scientific respectability. Indeed, reduc-
tionism as a general position can be seen as the answer provided by tho-
se who believe that an inclusive synthetic picture will never be achieved.
Science, as it has been developed during the past few centuries, has
no internal capacity to distinguish responsible from irresponsible pro-
jects. Scientists may be deeply responsible persons, or fathers/mothers,
or citizens, but there is no way internal to science to draw the boundary
between responsible and non responsible research projects and applica-
tions. This state of affairs can be seen as a consequence, possibly one of
the most important, of the overwhelming prevalence of analytic strate-
gies.
These factors notwithstanding, awareness is starting to spread that
something has gone wrong within mainstream science (and, let us add,
philosophy as well). As successful as analytic methods may be, at some
point they fail properly to grasp the phenomena under consideration. At
some point something different is required, something gentler and more
respectful of the integrity of the phenomena themselves.
The availability of both strategies (analytic and synthetic) will enab-
le the development of a more articulated, integral, respectful and respon-
sible vision of the world.
The synthetic vision required is anything but antiscientific. We de-
fend science in all its aspects. At the same time, however, we struggle
for a more integral science, a science able to integrate its many branches,
a science able to dialogue, to teach and to learn from both philosophy
and theology. This vision is headed toward a New Humanism, a spiritual
achievement that may be able to bridge many of the cleavages that seg-
ment our culture into isolated and isolating islands. As a side remark, it
may be worth noting that the trend toward more and more complex
forms of analytic specializations seems to have substantially contributed
to the general feeling that universities are failing badly in their roles.
The misplaced faith in analytic, reductionist methods follows almost

22
unavoidably from the lack of a correct ontology, and in particular from
the lack of both a theory of wholes and their parts and a theory of levels
of reality. More often than not, scholars defending otherwise widely dif-
ferent theories of spirituality have advocated some form or other of the
theory of levels (also known as the theory of aspects of reality). Thinkers
who have taken this approach include Husserl, Stein, Ingarden, Hart-
mann, Conrad-Martius and Dooyeweerd. All of them have realized that
without a properly developed theory of levels of reality, the deep, onto-
logical differences between organisms, minds and spirits become easily
blurred.
Following a somewhat different path, some scientists have also ad-
vocated the shift towards synthetic methods (Bohm and Rosen, among
others). Carl W. Hall, former deputy assistant director, directorate for
engineering, of the National Science Foundation (USA), began a paper
entitled “The Age of Synthesis” with the following words: “As we enter
the 21st
century we are in the beginning stages of a new age of synthesis,
philosophically and technologically. … This new age of synthesis is be-
ing built on and utilizes the emphasis on analysis featured in the 20th
ce-
ntury”. And then continued: “Synthesis provides a framework for gui-
ding analysis, research, development, management, and education”
[Hall, 1996, p. 12]. James F. Danielli, director of the Center for Theore-
tical Biology in Buffalo (1965-1974) and also the founder and long-time
editor of the Journal of Theoretical Biology (1962-1984) regularly used
the expression “age of synthesis”. Danielli was describing the three ages
in the science of modern biology as the age of observation → age of
analysis → age of synthesis. But the description is just as appropriate for
science in general.
2. Analysis and Synthesis
Analysis and synthesis have been used in many different ways. By
way of exemplification, the Greeks understood analysis as a working
back or reduction to principles, and Euclid masterfully applied it to geo-
metry. Synthesis then became the opposite strategy, which started from

23
principles and derived consequences from them. This initial understan-
ding of analysis and synthesis was also the starting point of a different
interpretation, according to which analysis is tied to induction and syn-
thesis to deduction. Closer to the present day, the fathers of analytic phi-
losophy, Frege and Russell, understood analysis as a transformative or
interpretive dimension. In order to apply the tools of logical analysis,
sentences should be transformed into their appropriate canonical form
([Dubray, n.d.], [Beaney, 2007]).
On the other hand, science uses analysis and synthesis in further dif-
ferent ways. As far as analysis is concerned, the term is used in fields as
different as spectrographic analysis, cost-benefit analysis and environ-
mental analysis, and many more besides. Synthesis, too, is used in wi-
dely different fields, such as chemical synthesis, frequency modulation
synthesis and Fourier synthesis, and many others.
Unfortunately, none of these more or less well documented under-
standings of analysis and synthesis fit well with the problems we are
about to discuss. For this reason, we shall propose a different interpre-
tation of analysis and synthesis implicitly sketched in the Introduction
above. One merit of the interpretation we are about to present is that it is
an interpretation grounded in the experience most of us have, even if it is
only rarely spelled out explicitly in all its necessary details.
According to the interpretation we are proposing, analysis and syn-
thesis are the two general strategies to which we may resort to answer
the basic question “what is this?” It is unquestionably true that there are
many different ways to answer this basic question. When I look at my
cat Sigliende and ask myself “what is Sigliende?” I may answer in a bio-
logically oriented fashion and say for instance: Sigliende is a cat, a feli-
ne, a mammal, an animal, a living being. Or I may instead answer by
stressing the connection between Sigliende and myself and say: Siglien-
de is a pet, or even better, Sigliende is my pet. These answers – and ma-
ny more – are unquestionably correct. Here, however, I would like to dig
deeper and try to understand what it is that makes Sigliende the entity
that it is. The “biologically oriented” kind of answer provides one way
of understanding the nature of Sigliende. A twin and possibly deeper
way to answer the same question is to adopt a structural framework, a
part-whole framework able to clarify which kinds of internal dependen-

24
cies are at work with Sigliende. In brief, this is precisely what most natu-
ral scientists do: they look for structures.
The question “what is this?” is the basis of any and every effort to
understand reality, be it the little, apparently irrelevant, item we are cu-
riously watching, or the marvel of a starry night or the sense of uncom-
fortable piety we feel at the suffering of a living being. Whatever it is,
whatever its value, we love and want to know.
As far as I can see, there are only four possible general strategies ab-
le to provide some answers to the “what is this?” question. Let us give a
name to “this” and call it “A”. To answer the question about the nature
of A we may adopt one of the following four strategies:
•A is what results from its parts (by way of a first approximation:
look downwards)
•A is what results from its neighbors (look sideways)
•A is what results from the wholes to which it pertains (look up-
wards)
•A is simply what it is, A is A and nothing else (look nowhere)
It is apparent that there is a major difference between the first three
answers, on the one hand, and the fourth answer, on the other. The first
three answers seem able to provide some information about the A we are
inquiring about, while the fourth answer refers to a formal constraint (the
identity of A), something apparently unable to provide any articulated
kind of information. It is worth noting that the formal reading provided
by the fourth answer above may be a necessary constraint required if our
mathematical and logical frameworks are to start working, and in this
sense it has indeed a remarkable role to play. Furthermore, it may have a
role to play as a general metaphysical principle by calling attention to
the fact that every individual entity is that particular individual, it has its
own individual nature, and for this reason it is different from any other
individual entity. This metaphysical side of the identity problem is usua-
lly called the haecceity of A, after a term introduced by the medieval
philosopher Duns Scotus. Both readings of identity, the formal and the
metaphysical ones, show that the fourth claim above is substantially mo-
re important than appears at first sight.
However, the main problem with the claim is that it is uninformati-

25
ve. It may play the role of a basic (formal or otherwise) constraint, and
in this sense it is something that any research framework should respect,
but in itself it does not provide further information. Even if Section 4 be-
low will show that there is still more to the matter, for the time being it
is plain that in order to articulate an informative answer to the question
“what is A?” we cannot but limit our consideration to one or more of the
first three answers.
The second answer (look sideways) also seems somewhat seconda-
ry. In order for some B that is not part of A to influence A there must be
a context C such that both A and B are part of C. Looking sideways is a
well-known piece of wisdom: more often than not, the behaviour of a
teenager is easier to understand by looking at her peers or at her parents,
as the situation may suggest. In either case, the answer is provided by
the whole that is the group of peers or the whole that is the family. Ans-
wer two is thus but a special case of answer three.
We are left with two main cases:
•A is what results from its parts (look downwards)
•A is what results from the wholes to which it pertains (look up-
wards)
Two more steps are needed to properly understand the workings of
analysis and synthesis. The next step is to consider that, more often than
not, in order to complete the answer to the initial question about the na-
ture of A, we must add a second question, namely “what does A do?”
This second question amounts to the imposition of constraints. In order
for item A to be able to do whatever it does, the parts of A and their inte-
ractions must be arranged in such a way that A can do whatever it does.
Complementarily, in order for the item A to be able to do whatever it do-
es, the wholes to which A pertains must be arranged in such a way that
A can do whatever it does. These two instances of making possible (a
concept I have learned from [Perzanowski, 1994]) play different roles:
from below making possible refers to the presence of enabling conditi-
ons, from above it refers to the lack of inhibiting conditions.
The final and main step is to set the window of inquiry. The main
option here is whether one single windowing is able to provide the infor-
mation we seek, or whether more (at least two) windowings are needed.

26
The three steps just sketched show that it is far too simple to conce-
ive analysis as only the strategy of looking downwards, and synthesis as
only the strategy of looking upwards. Analysis is instead the strategy of
looking downwards together with the assumption that one single windo-
wing of inquiry is sufficient. In order words, analysis claims that all the
information concerning item A are present in the framework composed
of the parts of A plus their mutual relations. By contrast synthesis starts
from the claim that more than one windowing of inquiry is needed. Let
us assume that two windowings are required, one for the whole A and
one for its parts. The claim that two windowings are needed makes sense
only if some of the information related to A is not present in its parts and
subsequent relations. In turn, this amounts to saying that what A does
imposes constraints on its parts and their relations. The parts are const-
rained precisely by the fact that they are parts of that whole. The prob-
lem is that the parts of A do not “see” the constraints imposed by A. In
other words, A’s parts must respect constraints they do not see. Analysis
has no way of answering the question of the constraints imposed from
above.
The distinction between single and multiple windowings brings to
light the connection between analysis and reductionism. If A can be enti-
rely understood by looking at the single windowing composed by its
parts plus their interactions, the same procedure can be repeated for the
parts themselves. We take them as wholes and entirely decompose them
into their own parts and interrelations. In the end we arrive at the rock-
bottom layer of reality, at what is really real, as they say. The fact that in
the meantime we lose all the rest of reality, ourselves included, does not
seem too worrying a conclusion for many scholars. I can perhaps grud-
gingly accept, at least for the time being, this unwelcome strategy, provi-
ded that no other options are available. But there is synthesis, and it
turns out that synthesis is both more informative and more general than
analysis. By admitting a number of windowings and their internal depen-
dences, the framework provided by synthesis can raise, and eventually
also try to answer, many more questions than any analytic framework
could ever dream of.

27
3. Wholes and Their Parts
Whichever aspect of reality interests us, it can be seen as, alternati-
vely, a whole or a part. Consider material reality: atoms form molecules,
which form cells, which form organisms. We can go further downwards
to subatomic particles or upwards to groups of organisms and ecosys-
tems. Leaving the two boundary cases aside, all the rest is a part when
seen from above and a whole when seen from below.
This is important, because it shows that as soon as one fixes the A
one is interested in, this A can be seen as either the whole that results
from its parts or as a part that plays a role within its whole(s). This stra-
tegy can be repeated for every and any kind of thing. Both strategies are
always available. This is indeed a rather fortunate situation. The possibi-
lity itself of science may indeed depend on the alternating play of wholes
and parts.
A subtle problem, however, emerges. Both strategies, as said, are al-
ways available, at least in principle. The problem, however, is that the
overall outcomes resulting from their systematic exploitation are remark-
ably different. One strategy pulls everything towards its lower constitu-
ents. As we have seen, analysis is the strategy adopted by all the many
versions of reductionism. On the other hand, the twin strategy of looking
upwards pulls everything towards some overall whole encompassing all
the aspects and nuances of reality. Untempered holism, however, can be
as utterly wrong as untempered reductionism. Both strategies, then, have
their dangers.
The discussion thus far, however, explains why universal analysis
down to the last parts and universal synthesis up to the most encompas-
sing whole are both unsupported extensions of the reasoning proposed
here. Both analysis and synthesis, in fact, depend on the previous move
of windowing item A, the unit under scrutiny. Analysis and synthesis are
always related to some A. It is with respect to A that we have to see
which of the two strategies is eventually more informative. In principle,
all options are available: analysis may prove more informative than syn-
thesis, or synthesis may prove more informative than analysis, or they
may be equally informative. It may also be that analysis proves more in-
formative for some aspects of the scrutiny, and synthesis for other as-

28
pects. Keeping item A firmly within the focus of scrutiny is therefore
mandatory to avoid the pitfalls of either untempered reductionism or un-
tempered holism.
4. The Three Main Kinds of Whole
As noted at the beginning of the paper, analysis tends to forget in-
formation. Analysis is blind to the information that is only valid at the
level of the whole. Unfortunately, there are cases where the most impor-
tant information characterizing the whole – what makes the whole what
it is – is precisely the information that analysis is unable to take into ac-
count. In order to spell out in some detail what I have just written, I need
to distinguish among three main types of wholes, namely simple, partial
and integral wholes.
Simple wholes are wholes that can be decomposed into parts with-
out losing information. Partial wholes are wholes that are not simple and
that require other wholes to exist and to do what they are supposed to do.
Integral wholes are wholes that are not simple and are maximal, in the
sense that they do not require other wholes of the same type to exist. For
them, doing usually implies further conditions, for reasons that will soon
become apparent.
The descriptive understanding of the various wholes can be sharpe-
ned by the following definitions, adopted after [Ingarden, 1964].
•Other-dependent: X is existentially other-dependent when the ex-
istence of X requires the existence of some other entity Y.
•Self-dependent: X is existentially self-dependent when X is not
other-dependent.
•Autonomous: X is existentially autonomous when X is other-de-
pendent and has its foundation in itself.
•Heteronomous = X is existentially heteronomous when X is other-
dependent and has its foundation in something else.
I shall soon say something on the foundation issue. Before that, the
three types of wholes that have been distinguished can then be specified
as follows:

29
•Simple wholes can be decomposed into parts without losing in-
formation.
•Partial wholes are not simple and are existentially heteronomous
wholes.
•Integral wholes are not simple and are existentially autonomous
wholes.
To better understand the difference between partial and integral
wholes, consider the difference between, say, an ear and an organism.
Both are wholes. Ears are authentic wholes, they can be studied by them-
selves in order to understand what they are and do. One can divide an
ear into its parts and see how they are made and what they do. The same
applies to an organism. Both are authentic wholes, both can be studied in
themselves, both can be subjected to (partial) analysis and synthesis. On
the other hand, it seems correct to claim that organisms are more comp-
letely wholes than ears.
Organisms have their own dependencies on other entities (air for
aerobic ones, food, mates, etc), because organisms are parts of higher-
order wholes, such as the ecological niche in which they live. All this is
patently true, and it amounts to saying that organisms are far from being
absolute wholes. On the other hand, it also seems correct to claim that
organisms are more important stop-points than ears, the latter being less
integral than the former.
Even if the idea of an integral whole is slippery and difficult to ma-
ke crystal-clear, it nevertheless appears to be both correct and quite natu-
ral. Reality has joints, and any proper understanding of reality should
learn to distinguish its joints. The fact that we are often wrong says more
about the limitations of our cognitive capacities than about reality itself.
4.1. Simple wholes
Simple wholes are the unquestioned realm of analysis. All questions
concerning simple wholes can be answered by decomposing them into
their parts and subsequent relations. Decomposition fully answers all the
questions that can be raised. Aggregates may be cases in point.

30
4.2. Partial wholes
Partial wholes are remarkably more interesting than simple wholes.
Consider the above example of the ear, which incidentally was the exa-
mple used in Riemann’s unfinished paper “The Mechanism of the Ear”
(1866) (see [Ritchey, 1996] for details). The starting point is the questi-
on: what is the ear supposed to do? It seems apparent that ears are such
that they make the organism able to perceive sounds. This answer has
two sides.
First, ears play some role within organisms. Like everything else,
they need wholes to be parts of. On the other hand, organisms build ears,
not the other way round. This provides some evidence for the claim that
ears are partial and not integral wholes.
Second, by having a role to play, ears cannot but constrain their own
parts in such a way that they end up forming a structure that is putatively
able to play the role it has to play. Parts should further constrain their
subparts so that they can play their own roles. And so on and so forth.
Two further observations are worth adding. The constraining proce-
dure may always fail. For any windowing, the constraints imposed by
the relevant whole may not be able to steer the appropriate developments
[Poli, An Introduction to the Ontology of Anticipation, 2009]. This may
imply, for instance, that the “ear” organ fails to play its role, and this in
its turn may further imply that the organism must either forget the perce-
ption of sounds or explore other avenues (e.g., different kinds of percep-
tion, or different kinds of ears). Secondly, the phenomenon of biological
convergence, according to which the most diverse biological species are
able to discover the same solutions, shows that, at this point, almost un-
known higher-order constraints are at work (see the astonishing evidence
accumulated by [Conway Morris, Life's Solution. Inevitable Humans in
a Lonely Universe, 2003] and [Conway Morris, The Deep Structure of
Biology, 2008]. See also [Poli, Evolution and Anticipation, 2010]).
When searching for a solution to its problems, life apparently does not
traverse the entire combinatorial space of possibilities, but continues to
discover the same solutions.

31
4.3. Integral Wholes
The case represented by integral wholes is substantially more intri-
guing than the previous two cases of simple and partial wholes. In fact,
integral wholes are wholes in a twofold way, so to speak. Not only are
they closer to what we have called above the joints of reality, but they
have further structural dimensions, lacking in the previous cases.
As a starter, the priority of organisms with respect to their organs
does not seem to pose too severe a problem. The issue is remarkably mo-
re intricate, however, where physical systems are concerned, because it
seems that there is no obvious way to segment physical items into integ-
ral as opposed to non-integral wholes or systems. Well, not really. Con-
sider the famous three-body problem. The three-body problem was – and
still is – one of the first theoretical failures of classical mechanics. In
brief, we can systematically calculate everything we need to calculate
with respect to the dynamics and mutual interactions of two bodies.
However, as soon as we pass from two to three interacting bodies, a sys-
tematic strategy is no longer available. Even if physicists and engineers
are able to calculate the numerical solution of each individual instance
of a three-body problem, no general analytic solution is available. The
reason is that the dynamics of a three-body system gives origin to an au-
thentic indecomposable whole, which implies that “one cannot solve a
three-body problem reductionistically, by solving two-body and one-bo-
dy problems” [Rosen, 2000, p. 109]. Any decomposition of a three-body
problem into a one-body and a two-body problem omits something (in
this case the integrality of their dynamics).
This simple example from baby-physics teaches us two lessons.
First, the integrality of an integral whole may be of a dynamic nature.
Second, the analysis-synthesis complementarity is even subtler than ex-
pected. In fact, we may be able to calculate any single individual instan-
ce of a given problem without this implying that we have a general me-
thod able to systematically address the problem. Sometimes we also
know that no general method will ever be available.
Some integral wholes exhibit the further intriguing structural feature
of producing their own parts. The capacity of a whole to produce the co-
mponents of which it is composed is usually called autopoiesis, after

32
[Maturana & Varela, 1980]. An autopoietic whole does not start from a
set of pre-given elements, nor does it assemble them. Autopoietic wholes
are self-referential systems, meaning that the whole’s relational self-pro-
duction governs the whole’s capacity to have contacts with its environ-
ment. Put otherwise, the whole’s connection with its environment beco-
mes a reflexive relation mediated by the self-referential loops that cons-
titute the whole itself. This single property changes the nature and work-
ings of the whole, dramatically strengthening the synthetic priority of the
whole with respect to its parts. Self-referentiality paves the way to the
third structural layer of this family of integral wholes, namely the fact
that these wholes may present a dynamic of their identity [Poli, The Co-
mplexity of Anticipation, 2009].
The definitions listed in section 4 above left the problem of ‘founda-
tion’ unaddressed. The classical understanding of foundation as referen-
ce to lower-level components obviously does not fit with the framework
we are articulating. Self-referential or autocatalytic loops may provide
an alternative worth pursuing. For more articulated details the reader
may consider [Louie, 2009], [Ulanowicz, 2009] and [Poli, The Comple-
xity of Anticipation, 2009].
In Section 2 above we set aside the case of identity as unable to pro-
vide information. We did not consider the case of those wholes that exist
in the form of changing entities. Besides the customary identity of a for-
mal type, these wholes also have some kind of dynamic identity. Orga-
nisms are cases in point; we are as well. Dynamic identity introduces an
entirely new class of integral wholes, namely wholes whose dynamics
may fulfill or fail to fulfill their potentialities. It is entirely appropriate
for these wholes to consider the problem of their flourishing, i.e. of their
capacity and possibility to unfold; one might say, to realize themselves.
These wholes have interests, things matter to them [Smith, 1998].
Thus far, we have exploited the framework of parts and wholes as
our sole framework of inquiry. However powerful and enlightening it
may well be, it may obscure some necessary distinctions. To glimpse
what the framework of wholes is not able to detect, one may ask oneself
whether atoms or molecules or cells can really be taken as parts of, say,
society. It obviously cannot be so. Therefore, something has been mis-
sed. To see what has been missed, let us move to the next framework,

33
the one provided by the theory of levels of reality.
5. Levels of Reality
The unity and the variety of the world is the outcome of the comp-
lex interweaving of dependence and inter-dependent connections and va-
rious forms of autonomy among the many items of which the world is
composed. The distinction is widespread among three basic realms or re-
gions (or strata, as we will call them) of reality. Even if the boundaries
between them are differently placed, the distinction between the three re-
alms of material, psychological and social phenomena is essentially ac-
cepted by most thinkers and scientists. A major source of discussion is
whether inanimate and animate beings should be placed in two different
realms (this meaning that there are in fact four and not three realms) or
within the same realm. The latter option defends the thesis that a phase
transition or something similar connects inanimate and animate items
(on the theory of levels of reality see [Poli, The Basic Problem of the
Theory of Levels of Reality, 2001], [Poli, First Steps in Experimental
Phenomenology, 2006], [Poli, Levels of Reality and the Psychological
Stratum, 2006], [Poli, Three Obstructions: Forms of Causation, Chrono-
topoids, and Levels of Reality, 2007].
Levels are distinguished by their categories. Material entities are di-
fferent from, say, psychological entities because their unfolding requires
different categorical frameworks. Leaving aside universal categories
(those that apply everywhere), two main categorical situations can be
distinguished:
(a)Types (Items) A and B are categorically different because the de-
scription (codification or modeling) of one of them requires cate-
gories that are not needed by the description (codification or mo-
deling) of the other.
(b)Types (Items) A and B are categorically different because their
description (codification or modeling) requires two entirely dif-
ferrent groups of categories.
I term the two relations as respectively relations of over-forming

34
and building-above. Strata or realms of reality are connected by buil-
ding-above relations. That is to say, the main reason for distinguishing
the different strata of reality as clearly as possible is that any one of them
is characterized by the birth of a new categorical series. The group of ca-
tegories that are needed to analyze the phenomena of the psychological
stratum is essentially different from the group of categories needed to
analyze the social one, which in its turn requires a group of categories
different from the one needed to analyze the material stratum of reality.
Over-forming (the type (a) form of categorical dependence) is wea-
ker than building-above and it is used to analyze aspects of the internal
organization of strata. Each of the three strata of reality has its specific
structure. The case of the material stratum is the best known and the le-
ast problematic. Suffice it to consider the series atom-molecule-cell-or-
ganism (which can be extended at each of its two extremes to include
sub-atomic particles and ecological communities, and also internally, as
needed). In this case we have a clear example of a series that proceeds
by levels of granularity. Compared to the material realm, the psycholo-
gical and social ones are characterized by an interruption in the material
categorical series and by the onset of new ones (relative to the psycholo-
gical and social items). More complex types of over-forming are instan-
tiated by them.
A terminological note may be helpful. I use the term ‘level’ to refer
in general to the levels of reality, restricting the term ‘layer’ to over-for-
ming relationships, and the term ‘stratum’ to building-above relation-
ships. I shall eventually use the expressions ‘sub-layer’ and ‘sub-stra-
tum’ when analysis requires them.
The question now arises as to how the material, psychological and
social strata are connected together. The most obvious answer is that
they have a linear structure. On this view, the social realm is founded on
the psychological stratum, which in its turn is founded on the material.
Likewise, the material stratum is the bearer of the psychological stratum,
which in its turn is the bearer of the social one. This point of view is part
of the received wisdom. However, a different option is possible. Consi-
der the possibility that material phenomena may act as bearers of both
psychological and social phenomena. In their turn, psychological and so-
cial phenomena reciprocally determine each other. Psychological and so-

35
cial systems are formed through co-evolution, meaning that the one is
the environmental prerequisite for the other [Luhmann, 1995].
This minimal introduction to the theory of levels is sufficient for our
purposes. In fact, what we really need is the relation we have called buil-
ding-above. What this shows is that the creativity of reality is such that it
is able to produce entities that are both existentially dependent on lower-
order entities and categorically orthogonal – i.e. entirely different – from
them. To reformulate what I have just said with less intimidating wor-
ding: minds and societies are entities deeply different from material
ones. However, we do not know of minds that flow away without their
organic hosts, nor have we ever, or ever will, encounter ungrounded so-
cieties, i.e. societies without individual agents.
How can it be that there are entities that are both dependent and or-
thogonal? We have already called attention to the most relevant cue to
begin answering such a demanding question. The cue was the remark
that it makes no sense to claim that atoms, molecules, or cells are parts
of societies, or minds for that matter. The fact that there is no proper
part-whole connection between them seems to ruin our entire perspecti-
ve. This is not so, however.
As a matter of fact, the impasse is more apparent than real, as shown
by our stress on the connection of existential dependence between lower-
order and higher-order entities. If these kinds of entities are connected
by a relation of existential dependence and are not related by a part-who-
le connection, the only remaining option is that they are linked by a
whole-whole relation. This latter kind of relation is much less-known
than the much better-known part-whole relations. Very few scholars ha-
ve been able to see the presence of whole-whole ties in the fabric of rea-
lity. The main reason for this widespread ‘blindness’ is that most thin-
kers do not have anything like a robust theory of levels of reality. The
only scholar who has chosen to delve into whole-whole relations is the
Dutch philosopher and theologian Herman Dooyeweerd, doing so main-
ly in his magnificent and impossible four-volume New Critique of Theo-
retical Thought ( [Dooyeweerd, 1953], for a lucid introduction to Doo-
yeweerd see [Clouser, 2005], for an application to biology [Zylstra,
1992]).

36
6. Whole-Whole Ties
Dooyeweerd distinguished different types of enkapsis – his name
for whole-whole ties – namely
•Foundational enkapsis, such as the sculpture, and the block of mar-
ble from which it is made.
•Subject-object enkapsis, such as a hermit crab and its shell.
•Symbiotic enkapsis, such as clover and its nitrogen-fixing bacteria.
•Correlative enkapsis, such as an environment and its denizens.
•Territorial enkapsis, such as a city and its university [Basden,
2008, p. 89].
In this paper, I shall restrict my remarks to the first type of enkapsis,
the so-called foundational enkapsis. To grasp the framework addressed
by foundational enkapsis more clearly, let us consider a few relevant ca-
ses, such as those exemplified by the following ties:
•The marble—statue tie
•The canvas—painting tie
•The paper—water-colour tie
•The paper—novel tie
•The CD—song tie
To simplify our inquiry we have chosen cases pertaining to the same
sub-family of foundational enkapsis, namely the family of works of art.
It is apparent that all the above five cases show that there is a connection
between something that behaves as a bearer and something else that is
borne by it. Furthermore, the features that describe the nature of the ob-
jects playing the role of bearer and the objects that are borne by them are
broadly if not entirely different. The physical properties of marble, can-
vas, paper and a CD, in fact, are remarkably different from the aesthetic
properties of the statue, painting, water-colour novel and song. Let us
call these properties the ‘nature’ or the ‘qualifying function’ of the res-
pective objects.
On deeper analysis, however, an interesting subtlety surfaces. The

37
five exemplifications above can be divided into two different groups dis-
tinguished by whether the bearer has some interaction with the object
that it bears. CDs and the paper used to print a novel, in fact, have no
kind of interaction with the higher-order objects they bear, as proved by
the fact that the latter objects can be just as effectively borne by other
bearers, such as mp3 or pdf files. Electronic versions of novels and
songs are as authentic as paper-printed or CD-printed versions.
On the other hand, however, the tie between a water-colour and the
paper on which it is painted is more intimate, because the colour penet-
rates into the paper’s fibers. It is well known, in fact, that water-colours
should be painted on special kinds of paper which let the colour penetra-
te into their fibers because this adds further layers of expressivity to the
painting. Similarly, not all types of marble are equally suitable for a gi-
ven statue, and the properties of the marble add something to the aesthe-
tic properties of the statue.
As far as the interplay between an aesthetic object and its bearer is
concerned, the canvas-painting case is somewhat intermediate between
the two groups thus far discussed, in the sense that it is far less evident
whether the bearer influences the object it bears. Another characteristic,
however, places this latter case closer to the second group. The two gro-
ups distinguished can be seen from a different perspective, namely whe-
ther they are truly reproducible. The exemplifications belonging to the
first group (songs and novels) can be reproduced as wanted, and all of
them remain true exemplifications of the same object. This further exp-
lains why the bearer is utterly irrelevant to the borne object: some bearer
is needed in order to instantiate the object, but what kind of bearer is
used is utterly immaterial. On the other, hand, the other group is compo-
sed of objects that cannot be truly reproduced, in the sense that properly
speaking any reproduction is a fake. It may be a clone, but it is neverthe-
less a fake: it is a different object from the original. The non-reproduci-
bility of this family of objects patently depends on the more intimate co-
nnection between the borne component and its bearer. As before, the
canvas-painting case is somewhat intermediate between the two main
cases, in the sense that a painting can be transferred to a different canvas
(e.g. when the original is so damaged that the entire painting risks dest-
ruction). We can leave further complications aside, however, since we

38
have already accumulated enough data to understand foundational enka-
psis.
The analysis has thus far shown that there are at least some objects
with a stratified structure organized in such a way that their strata are
linked by a double connection: first, the higher stratum existentially de-
pends on its lower stratum (it must be instantiated into some “matter”)
and, second, the properties of the two strata are widely different if not
utterly orthogonal. The further distinction between reproducible and non
reproducible instances shows that other components may have to be ta-
ken into account.
The colours used by an artist are themselves material entities – and
this explains why they can interact with the material surfaces on which
they are placed. The features of the marble are explicitly exploited by
the artist when she gives shape to her work; they are information that en-
ters the fabric of the work of art.
The five cases we have seen are far from exemplifying authentic
part-whole relations, because the two strata of the bearer and the borne
have different natures. An authentic part-whole relation can only work
between objects with the same nature (or qualifying function as Dooye-
weerd would say [Clouser, 2005, p. 286]). The existential dependence of
the higher stratum on the lower is thus far from being a sufficient condi-
tion for a part-whole relation. We need air to keep ourselves alive, we
existentially depend on it, but air is not one of our parts.
I shall follow [Clouser, 2005] in describing the greater wholes
which include sub-wholes as “encapsulating” them, and thus call the
greater whole “capsulate wholes”. Capsulation stands for the awkward
enkapsis.
Capsulate wholes are everywhere. Molecules capsulate atoms, and
cells capsulate molecules, and so on and so forth. Works of art capsulate
their bearers. For all these cases, the nature (or qualifying function) of
the capsulate whole overrides the nature of its capsulated sub-wholes
[Clouser, 2005, p. 289].

39
7. Wholes again
It is worth noting that the interest surrounding the analysis of parts
and wholes, so popular in the past 20 years or so is almost completely
focused on the relation “part-of”. The non-relational category of whole,
on the other hand, has been far less frequently addressed, apparently for
a number of good reasons. Not only are wholes more refractory to cate-
gorical scrutiny, but the viewpoint of wholes has historically been con-
nected to visions that today do not have much currency, such as the the-
ory of the so-called ethical state developed by Hegel and other idealist
thinkers. It is also well-known that some of the most obviously outdated
proposals advanced by Aristotle, notably his theories on the state and the
family, and the consequent subordinate role of women and slaves, direc-
tly depended on his view of the state and the family as wholes.
The most straightforward reading of these issues can be aptly sum-
marized thus: the family and the state are wholes, the husband/father/
king is the formal representative of the family (“is” the family) or the
state (“is” the state), and every other member of the family/state must be
subordinated to him/it.
However, this reading is critically dependent on a specific assump-
tion; namely that a natural whole always has (must have) a canonical re-
presentative. The issue is a rather subtle one and involves a number of
problems we cannot discuss here (see [Poli, The Complexity of Antici-
pation, 2009] for some of the connected complications). Leaving many
details aside, the main question is nevertheless apparent. Why should a
whole have one unique individual representative? Only specific – i.e.
non-generic – wholes do. In particular, only (totally, i.e. perfectly) hie-
rarchical wholes have maxima. This shows that the theory of wholes ca-
nnot be restricted to those wholes that have canonical representatives.
One of the distinctive characteristics of modern society – as oppo-
sed to traditional forms of society – is precisely the transformation from
an essentially hierarchical structure (well represented by the king) to a
functional organization in which politics, law, economics, art, religion,
science, and many more dimensions each has its own irreplaceable role
to play. There is no natural way to confine functional structures within
one single total hierarchy. The same applies to the family: both spouses

40
are equally representatives of the whole that is their family.
Hierarchical wholes are then but a tiny subsection of wholes, and it
is simply wrong to conceive wholes as hierarchical. Some are, many mo-
re are not.
8. Conclusion
This paper has begun to unfold the multi-layered complexity of the
connections linking together analysis and synthesis. Even if much work
has still to be done, the overall geography of the problems starts to exhi-
bit a few landmarks which should be exploited further. Three appear
worth mentioning in the paper’s final summary, namely:
•The ontological relevance of windowing;
•The typology of wholes, including their relationships and in parti-
cular the whole-whole family of connections; and
•Levels of reality or what can be termed an ontological theory of ty-
pes.
These three headings, if addressed, may eventually contribute to the
development of a more articulate ontology, capable of grasping deeper-
lying aspects of reality.
References
Basden, A. (2008). Philosophical Frameworks for Understanding Infor-
mation Systems. Hershey, New York: IGI Publishing.
Beaney, M. (2007). Analysis. Retrieved November 2008, from Stanford
Encyclopedia of Philosophy: http://plato.stanford.edu/entries/analysis/
Clouser, R. (2005). The Myth of Religious Neutrality. An Essay on the
Hidden Role of Religious Belief in Theories, 2nd Revised edition. Notre
Dame: University of Notre Dame Press.
Conway Morris, S. (2003). Life's Solution. Inevitable Humans in a Lone-
ly Universe. Cambridge: Cambridge University Press.

41
Conway Morris, S. (2008). The Deep Structure of Biology. West Con-
shohocken (Pennsylvania): Templeton Foundation Press.
Dooyeweerd, H. (1953). A New Critique of Theoretical Thought . Phila-
delphia: The Presbyterian and Reformed Publishing Company.
Dubray, C. (n.d.). Analysis. Retrieved November 2008, from New Ad-
vent (The Catholic Encyclopedia. New York: Robert Appleton Compa-
ny, 1907): http://www.newadvent.org/cathen/01450a.htm
Hall, C. W. (1996). The Age of Synthesis. Resource , 12-14.
Ingarden, R. (1964). Time and modes of being. Springfield: Charles C.
Thomas Publisher.
Louie, A. (2009). More Than Life Itself. Frankfurt: Ontos Verlag.
Luhmann, N. (1995). Social Systems. Stanford: Stanford University
Press.
Maturana, H., & Varela, F. (1980). Autopoiesis and Cognition. Boston:
Reidel.
Perzanowski, J. (1994). Reasons and Causes. In J. Faye, U. Scheffler, &
M. Urchs, Logic and Causal Reasoning (pp. 169-189). Berlin: Akademie
Verlag.
Poli, R. (2010). Evolution and Anticipation. submitted.
Poli, R. (2009). An Introduction to the Ontology of Anticipation. Futu-
res, forthcoming.
Poli, R. (2009). The Complexity of Anticipation. Balkan Journal of Phi-
losophy, 1 (1), 19-29.
Poli, R. (2007). Three Obstructions: Forms of Causation, Chronotopoids,
and Levels of Reality. Axiomathes , 1-18.
Poli, R. (2006). First Steps in Experimental Phenomenology. In A. Lou-
la, R. Gudwin, & J. Queiroz, Artificial Cognition Systems (pp. 358-386).
Hersey PA: Idea Group Publishing.

42
Poli, R. (2006). Levels of Reality and the Psychological Stratum. Revue
Internationale de Philosophie, 163-180.
Poli, R. (2001). The Basic Problem of the Theory of Levels of Reality.
Axiomathes, 12 (3-4), 261-283.
Ritchey, T. (1996). Analysis and Synthesis. Retrieved November 2008,
from Swedish Morphological Society: http://www.swemorph.com
Rosen, R. (2000). Essays on Life Itself. New York: Columbia University
Press.
Smith, B. C. (1998). God, Approximately. Retrieved November 2008 ,
from
http://www.ageofsignificance.org/people/bcsmith/papers/index.html
Ulanowicz, R. E. (2009). A Third Window. Natural Life beyond Newton
and Darwin . West Conshohocken: Templeton Foundation Press.
Zylstra, U. (1992). Living things as hierarchically organized structures.
Synthese , 111-133.

Analysis-Synthesis

More Related Content

Similar to Analysis-Synthesis

More from Kimberly Pulley

Recently uploaded

Analysis-Synthesis