of the garden included in his field of view (our operational definition of a garden unit)
In practice the Observer will be usually moving along a certain path so that under his field of view new gardening units will be appearing all the time. The messages received
in this situation will be organized or integrated at his brain and the problem concerning us now is how to adapt the models proposed to a condition of chainging visual inputs.
A garden unit (GU) as defined, constitutes one component of a larger picture or
scene;this picture may be said to be made up by the contributions of all the GUs
included in it. However, the larger picture will most unlikely be the result of the linear
summation of the individual GUs. in the way its total biomass will be the sum of those
of the components. A masterful discusion of the problem of landscapes was presented by George Santayana in his treatise on Aesthetics [ The Sense of Beauty;Dover Publ,
(1955)]. Our arguments against the linear summation are a) that the boundaries of the GUs are not only fixed in space but also in time b) that the attention given to different sections will undoubtedly not be the same and c) the memory registration of
the scene will result from a) and b).
As a result of the above, we could safely assume that the GUs are hierarchically organized (as opposed to linearly organized) with the levels
determined mainly by continuity and degree of attention.
A set of a given number of contiguous GUs is a sub-set of the whole Garden but it is in itself made up by the number of individual GUs. Each GU, in turn,
is made up of a number of plant components and so forth. The term Holarchy has
been proposed to describe a whole that is also a structure of parts that are in themselves
whole. In a Holarchy each of these components constitutes a holon, a term initially
proposed by Arthur Koestler in an admirable book called The Ghost in the Machine;
Hutchinson & Co.(1967). See also holonic solutions for more info.
Holons, according to Koestler refer to " any stable biological or social sub-whole
which displays rule governed behaviour and/or structural Gestalt-constancy".
[ The subject of whether or not the Gardening Units proposed in this work
may be considered holons is discussed in the section Gardens as Holarchies. There also
the use of environs, proposed by the ecologist B.C. Patten, for the system's environment is considered ( Patten: The Am. Naturalist,119, 179-218, 1982) ]
When dealing with the organization of visual perceptions from
a number of gardening units, a particularly suitable parallel is one (also borrowed
from Koestler) dealing with the written word: In a book, letters are organized as words,
words organized as sentences, sentences in paragraphs, paragraphs in chapters and so
forth. The parallel being the book as the whole garden, chapters as garden sections,
paragraphs as garden blocks(sets of adjacent GUS) sentences as garden units,words as
plants and letters as plant parts. To say that a paragraph is the sum of all the sentences
included in it is not a false statement but it is just a poor description. When we try to recall a given paragraph only one or two sentences will come back, the ones expressing
ideas which for one reason or other have impressed our brain/mind. Most of the
other sentences will be sort of blurred and will be recalled, if at all, with great effort.
Now, to say that a garden block is the sum of the garden units included is also not
false but just a poor description from the Observer point of view. If he tries
to recall the overall picture, some salient features (pleasant or unpleasant) will come
back, the rest being just a background more or less blurred.
In Diagram 6, below, I have tried to illlustrate in a simplified form the problem of
the organization by the Observer of a set of contigous gardening units. Let's assume the observer walks along a path, at a slow pace and paying attention to the garden at his left side. Whenever he stops, his field of view can encompass a gardening unit consisting of
four cells (say, 1 to 4, or 5 to 8). These GUs are designated a,b,c...etc., or ab, bc,cd...
etc., depending on which celles he happens to be looking for a given time interval.
( Note: this is a discontinous or step by step process; we can approach a continous process by increasing the number of cells for each GU, but I feel that at this stage this
further complication is not justified).
Let's further suppose that the light green cells of Diagram 6 represent areas covered by a cover plant, say trailing myoporon not in bloom, of uniform colour and
unremarkable texture, in short, a uniform green carpet. Suddenly the Observer arrives at "d" and encounters a huge bouganvillea in full bloom and continuing further, he encounters the same cover plant as before.
a SNR of 90, then the observer's brain receive visual messages of intensities as those
depicted on diagram 7 as he proceeds along the path. This is as far as we can get with our adaptation of Shannon and Weaber's model. Of more importance is the "judgement"
by the observer at the end of the path. Our contention, within the context of Clause's
model is that his recollection of the garden will be primarily influenced by the message
from cells 13 to 16, notwithstanding the fact that they constitute less than 20% of the
garden area and 10% of the residence time.
garden block is certaily not the average of those of the component GUs but a value
close to that of the dominant GU. Along similar lines it may be proposed that if a
disturbance with a very low SNR is present, then the value for the whole garden block
will be close to that of the disturbance. If in Diagram 6 the cells 13 to 16 are replaced
by a patch of dry and badly tramped myoporon with SNR of say, five, then the overall value will be close to 5.
