Differential Propositional Calculus
A differential propositional calculus is a propositional calculus extended by a set of terms for describing processes that take place in a universe of discourse or transformations that map a source universe into a target universe.
Casual introduction
…
Formal development
…
Expository examples
…
Differential Logic : Series A
Differential Propositions
One of the first things that you can do, once you have a really decent calculus for boolean functions or propositional logic, whatever you want to call it, is to compute the differentials of these functions or propositions.
Now there are many ways to dance around this idea, and I feel like I have tried them all, before one gets down to acting on it, and there many issues of interpretation and justification that we will have to clear up after the fact, that is, before we can be sure that it all really makes any sense, but I think this time I'll just jump in, and show you the form in which this idea first came to me.
Start with a proposition of the form x & y, which I graph as two labels attached to a root node, so:
o-------------------------------------------------o | | | x y | | @ | | | o-------------------------------------------------o | x and y | o-------------------------------------------------o
Written as a string, this is just the concatenation x y.
The proposition xy may be taken as a boolean function f(x, y) having the abstract type f : B × B → B, where B = {0, 1} is read in such a way that 0 means false and 1 means true.
In this style of graphical representation, the value true looks like a blank label and the value false looks like an edge.
o-------------------------------------------------o | | | | | @ | | | o-------------------------------------------------o | true | o-------------------------------------------------o
o-------------------------------------------------o | | | o | | | | | @ | | | o-------------------------------------------------o | false | o-------------------------------------------------o
Back to the proposition xy. Imagine yourself standing in a fixed cell of the corresponding venn diagram, say, the cell where the proposition xy is true, as pictured:
o-------------------------------------------------o | | | | | o-----------o o-----------o | | / \ / \ | | / o \ | | / /%\ \ | | / /%%%\ \ | | o o%%%%%o o | | | |%%%%%| | | | | |%%%%%| | | | | x |%%%%%| y | | | | |%%%%%| | | | | |%%%%%| | | | o o%%%%%o o | | \ \%%%/ / | | \ \%/ / | | \ o / | | \ / \ / | | o-----------o o-----------o | | | | | o-------------------------------------------------o
Now ask yourself: What is the value of the proposition xy at a distance of dx and dy from the cell xy where you are standing?
Don't think about it -- just compute:
o-------------------------------------------------o | | | dx o o dy | | / \ / \ | | x o---@---o y | | | o-------------------------------------------------o | (x + dx) and (y + dy) | o-------------------------------------------------o
To make future graphs easier to draw in Ascii land, I will use devices like @=@=@ and o=o=o to identify several nodes into one, as in this next redrawing:
o-------------------------------------------------o | | | x dx y dy | | o---o o---o | | \ | | / | | \ | | / | | \| |/ | | @=@ | | | o-------------------------------------------------o | (x + dx) and (y + dy) | o-------------------------------------------------o
However you draw it, these expressions follow because the expression x + dx, where the plus sign indicates (mod 2) addition in B, and thus corresponds to an exclusive-or in logic, parses to a graph of the following form:
o-------------------------------------------------o | | | x dx | | o---o | | \ / | | @ | | | o-------------------------------------------------o | x + dx | o-------------------------------------------------o
Next question: What is the difference between the value of the proposition xy "over there" and the value of the proposition xy where you are, all expressed as general formula, of course? Here 'tis:
o-------------------------------------------------o | | | x dx y dy | | o---o o---o | | \ | | / | | \ | | / | | \| |/ x y | | o=o-----------o | | \ / | | \ / | | \ / | | \ / | | \ / | | \ / | | @ | | | o-------------------------------------------------o | ((x + dx) & (y + dy)) - xy | o-------------------------------------------------o
Oh, I forgot to mention: Computed over B, plus and minus are the very same operation. This will make the relationship between the differential and the integral parts of the resulting calculus slightly stranger than usual, but never mind that now.
Last question, for now: What is the value of this expression from your current standpoint, that is, evaluated at the point where xy is true? Well, substituting 1 for x and 1 for y in the graph amounts to the same thing as erasing those labels:
o-------------------------------------------------o | | | dx dy | | o---o o---o | | \ | | / | | \ | | / | | \| |/ | | o=o-----------o | | \ / | | \ / | | \ / | | \ / | | \ / | | \ / | | @ | | | o-------------------------------------------------o | ((1 + dx) & (1 + dy)) - 1&1 | o-------------------------------------------------o
And this is equivalent to the following graph:
o-------------------------------------------------o | | | dx dy | | o o | | \ / | | o | | | | | @ | | | o-------------------------------------------------o | dx or dy | o-------------------------------------------------o
We have just met with the fact that the differential of the "and" is the "or" of the differentials.
- x and y --Diff--> dx or dy.
o-------------------------------------------------o | | | dx dy | | o o | | \ / | | o | | x y | | | @ --Diff--> @ | | | o-------------------------------------------------o | x y --Diff--> ((dx) (dy)) | o-------------------------------------------------o
It will be necessary to develop a more refined analysis of this statement directly, but that is roughly the nub of it.
If the form of the above statement reminds you of De Morgan's rule, it is no accident, as differentiation and negation turn out to be closely related operations. Indeed, one can find discussions of logical difference calculus in the Boole-De Morgan correspondence and C.S. Peirce also made use of differential operators in a logical context, but the exploration of these ideas has been hampered by a number of factors, not the least of which being a syntax adequate to handle the complexity of expressions that evolve.
For my part, it was definitely a case of the calculus being smarter than the calculator thereof. The graphical pictures were catalytic in their power over my thinking process, leading me so quickly past so many obstructions that I did not have time to think about all of the difficulties that would otherwise have inhibited the derivation. It did eventually became necessary to write all this up in a linear script, and to deal with the various problems of interpretation and justification that I could imagine, but that took another 120 pages, and so, if you don't like this intuitive approach, then let that be your sufficient notice.
Let us run through the initial example again, this time attempting to interpret the formulas that develop at each stage along the way.
We begin with a proposition or a boolean function f(x, y) = xy.
o-------------------------------------------------o | | | | | o-----------o o-----------o | | / \ / \ | | / o \ | | / /`\ \ | | / /```\ \ | | o o`````o o | | | |`````| | | | | |`````| | | | | x |``f``| y | | | | |`````| | | | | |`````| | | | o o`````o o | | \ \```/ / | | \ \`/ / | | \ o / | | \ / \ / | | o-----------o o-----------o | | | | | o-------------------------------------------------o | | | x y | | @ | | | o-------------------------------------------------o | f = x y | o-------------------------------------------------o
A function like this has an abstract type and a concrete type. The abstract type is what we invoke when we write things like f : B × B → B or f : B2 → B. The concrete type takes into account the qualitative dimensions or the "units" of the case, which can be explained as follows.
- Let X be the set of values {(x), x} = {not x, x}.
- Let Y be the set of values {(y), y} = {not y, y}.
Then interpret the usual propositions about x, y as functions of the concrete type f : X × Y → B.
We are going to consider various "operators" on these functions. Here, an operator F is a function that takes one function f into another function Ff.
The first couple of operators that we need to consider are logical analogues of those that occur in the classical "finite difference calculus", namely:
- The difference operator Δ, written here as D.
- The enlargement operator Ε, written here as E.
These days, E is more often called the shift operator.
In order to describe the universe in which these operators operate, it will be necessary to enlarge our original universe of discourse. We mount up from the space U = X × Y to its differential extension, EU = U × dU = X × Y × dX × dY, with dX = {(dx), dx} and dY = {(dy), dy}. The interpretations of these new symbols can be diverse, but the easiest for now is just to say that dx means "change x" and dy means "change y". To draw the differential extension EU of our present universe U = X × Y as a venn diagram, it would take us four logical dimensions X, Y, dX, dY, but we can project a suggestion of what it's about on the universe X × Y by drawing arrows that cross designated borders, labeling the arrows as dx when crossing the border between x and (x) and as dy when crossing the border between y and (y), in either direction, in either case.
o-------------------------------------------------o | | | | | o-----------o o-----------o | | / \ / \ | | / x o y \ | | / /`\ \ | | / /```\ \ | | o o`````o o | | | |`````| | | | | dy |`````| dx | | | | <---------|--o--|---------> | | | | |`````| | | | | |`````| | | | o o`````o o | | \ \```/ / | | \ \`/ / | | \ o / | | \ / \ / | | o-----------o o-----------o | | | | | o-------------------------------------------------o
We can form propositions from these differential variables in the same way that we would any other logical variables, for instance, interpreting the proposition (dx (dy)) to say "dx ⇒ dy", in other words, however you wish to take it, whether indicatively or injunctively, as saying something to the effect that there is "no change in x without a change in y".
Given the proposition f(x, y) in U = X × Y, the (first order) enlargement of f is the proposition Ef in EU that is defined by the formula Ef(x, y, dx, dy) = f(x + dx, y + dy).
In the example f(x, y) = xy, we obtain:
- Ef(x, y, dx, dy) = (x + dx)(y + dy).
o-------------------------------------------------o | | | x dx y dy | | o---o o---o | | \ | | / | | \ | | / | | \| |/ | | @=@ | | | o-------------------------------------------------o | Ef = (x, dx) (y, dy) | o-------------------------------------------------o
Given the proposition f(x, y) in U = X × Y, the (first order) difference of f is the proposition Df in EU that is defined by the formula Df = Ef – f, or, written out in full, Df(x, y, dx, dy) = f(x + dx, y + dy) – f(x, y).
In the example f(x, y) = xy, the result is:
- Df(x, y, dx, dy) = (x + dx)(y + dy) – xy.
o-------------------------------------------------o | | | x dx y dy | | o---o o---o | | \ | | / | | \ | | / | | \| |/ x y | | o=o-----------o | | \ / | | \ / | | \ / | | \ / | | \ / | | \ / | | @ | | | o-------------------------------------------------o | Df = ((x, dx)(y, dy), xy) | o-------------------------------------------------o
We did not yet go through the trouble to interpret this (first order) difference of conjunction fully, but were happy simply to evaluate it with respect to a single location in the universe of discourse, namely, at the point picked out by the singular proposition xy, in as much as if to say, at the place where x = 1 and y = 1. This evaluation is written in the form Df|xy or Df|<1, 1>, and we arrived at the locally applicable law that states that f = xy = x & y ⇒ Df|xy = ((dx)(dy)) = dx or dy.
o-------------------------------------------------o | | | | | o-----------o o-----------o | | / \ / \ | | / x o y \ | | / /`\ \ | | / /```\ \ | | o o`````o o | | | |`````| | | | | dy (dx) |`````| dx (dy) | | | | o<----------|--o--|---------->o | | | | |``|``| | | | | |``|``| | | | o o``|``o o | | \ \`|`/ / | | \ \|/ / | | \ | / | | \ /|\ / | | o-----------o | o-----------o | | | | | dx|dy | | | | | v | | o | | | o-------------------------------------------------o | | | dx dy | | o o | | \ / | | o | | | | | @ | | | o-------------------------------------------------o | Df|xy = ((dx) (dy)) | o-------------------------------------------------o
The picture illustrates the analysis of the inclusive disjunction ((dx)(dy)) into the exclusive disjunction: dx(dy) + dy(dx) + dx dy, a proposition that may be interpreted to say "change x or change y or both". And this can be recognized as just what you need to do if you happen to find yourself in the center cell and desire a detailed description of ways to depart it.
We have just computed what will variously be called the difference map, the difference proposition, or the local proposition Dfp for the proposition f(x, y) = xy at the point p where x = 1 and y = 1.
In the universe U = X × Y, the four propositions xy, x(y), (x)y, (x)(y) that indicate the "cells", or the smallest regions of the venn diagram, are called singular propositions. These serve as an alternative notation for naming the points <1, 1>, <1, 0>, <0, 1>, <0, 0>, respectively.
Thus, we can write Dfp = Df|p = Df|<1, 1> = Df|xy, so long as we know the frame of reference in force.
Sticking with the example f(x, y) = xy, let us compute the value of the difference proposition Df at all of the points.
o-------------------------------------------------o | | | x dx y dy | | o---o o---o | | \ | | / | | \ | | / | | \| |/ x y | | o=o-----------o | | \ / | | \ / | | \ / | | \ / | | \ / | | \ / | | @ | | | o-------------------------------------------------o | Df = ((x, dx)(y, dy), xy) | o-------------------------------------------------o
o-------------------------------------------------o | | | dx dy | | o---o o---o | | \ | | / | | \ | | / | | \| |/ | | o=o-----------o | | \ / | | \ / | | \ / | | \ / | | \ / | | \ / | | @ | | | o-------------------------------------------------o | Df|xy = ((dx) (dy)) | o-------------------------------------------------o
o-------------------------------------------------o | | | o | | dx | dy | | o---o o---o | | \ | | / | | \ | | / o | | \| |/ | | | o=o-----------o | | \ / | | \ / | | \ / | | \ / | | \ / | | \ / | | @ | | | o-------------------------------------------------o | Df|x(y) = (dx) dy | o-------------------------------------------------o
o-------------------------------------------------o | | | o | | | dx dy | | o---o o---o | | \ | | / | | \ | | / o | | \| |/ | | | o=o-----------o | | \ / | | \ / | | \ / | | \ / | | \ / | | \ / | | @ | | | o-------------------------------------------------o | Df|(x)y = dx (dy) | o-------------------------------------------------o
o-------------------------------------------------o | | | o o | | | dx | dy | | o---o o---o | | \ | | / | | \ | | / o o | | \| |/ \ / | | o=o-----------o | | \ / | | \ / | | \ / | | \ / | | \ / | | \ / | | @ | | | o-------------------------------------------------o | Df|(x)(y) = dx dy | o-------------------------------------------------o
The easy way to visualize the values of these graphical expressions is just to notice the following equivalents:
o-------------------------------------------------o | | | x | | o-o-o-...-o-o-o | | \ / | | \ / | | \ / | | \ / x | | \ / o | | \ / | | | @ = @ | | | o-------------------------------------------------o | (x, , ... , , ) = (x) | o-------------------------------------------------o
o-------------------------------------------------o | | | o | | x_1 x_2 x_k | | | o---o-...-o---o | | \ / | | \ / | | \ / | | \ / | | \ / | | \ / x_1 ... x_k | | @ = @ | | | o-------------------------------------------------o | (x_1, ..., x_k, ()) = x_1 ... x_k | o-------------------------------------------------o
Laying out the arrows on the augmented venn diagram, one gets a picture of a differential vector field.
o-------------------------------------------------o | | | o | | | | | dx|dy | | | | | o-----------o | o-----------o | | / \|/ \ | | / x | y \ | | / /|\ \ | | / /`|`\ \ | | o o``|``o o | | | dy (dx) |``v``| dx (dy) | | | | o-----------|->o<-|-----------o | | | | |`````| | | | | o<----------|--o--|---------->o | | | | dy (dx) |``|``| dx (dy) | | | o o``|``o o | | \ \`|`/ / | | \ \|/ / | | \ | / | | \ /|\ / | | o-----------o | o-----------o | | | | | dx|dy | | | | | v | | o | | | o-------------------------------------------------o
This really just constitutes a depiction of the interpretations in EU = X × Y × dX × dY that satisfy the difference proposition Df, namely, these:
1. x y dx dy 2. x y dx (dy) 3. x y (dx) dy 4. x (y)(dx) dy 5. (x) y dx (dy) 6. (x)(y) dx dy
By inspection, it is fairly easy to understand Df as telling you what you have to do from each point of U in order to change the value borne by f(x, y).
We have been studying the action of the difference operator D, also known as the localization operator, on the proposition f : X × Y → B that is commonly known as the conjunction xy. We described Df as a (first order) differential proposition, that is, a proposition of the type Df : X × Y × dX × dY → B. Abstracting from the augmented venn diagram that illustrates how the models, or the satisfying interpretations, of Df distribute within the extended universe EU = X × Y × dX × dY, we can depict Df in the form of a digraph or directed graph, one whose points are labeled with the elements of U = X × Y and whose arrows are labeled with the elements of dU = dX × dY.
o-------------------------------------------------o | f = x y | o-------------------------------------------------o | | | Df = x y ((dx)(dy)) | | | | + x (y) (dx) dy | | | | + (x) y dx (dy) | | | | + (x)(y) dx dy | | | o-------------------------------------------------o | | | x y | | x (y) o<------------->o<------------->o (x) y | | (dx) dy ^ dx (dy) | | | | | | | | dx | dy | | | | | | | | v | | o | | (x) (y) | | | o-------------------------------------------------o
Any proposition worth its salt has many equivalent ways to view it, any one of which may reveal some unsuspected aspect of its meaning. We will encounter more and more of these variant readings as we go.
The enlargement operator E, also known as the shift operator, has many interesting and very useful properties in its own right, so let us not fail to observe a few of the more salient features that play out on the surface of our simple example, f(x, y) = xy.
Introduce a suitably generic definition of the extended universe of discourse:
- Let U = X1 × … × Xk and EU = U × dU = X1 × … × Xk × dX1 × … × dXk.
For a proposition f : X1 × … × Xk → B, the (first order) enlargement of f is the proposition Ef : EU → B that is defined by:
- Ef(x1, …, xk, dx1, …, dxk) = f(x1 + dx1, …, xk + dxk).
It should be noted that the so-called differential variables dxj are really just the same kind of boolean variables as the other xj. It is conventional to give the additional variables these brands of inflected names, but whatever extra connotations we might choose to attach to these syntactic conveniences are wholly external to their purely algebraic meanings.
For the example f(x, y) = xy, we obtain:
- Ef(x, y, dx, dy) = (x + dx)(y + dy).
Given that this expression uses nothing more than the boolean ring operations of addition (+) and multiplication (·), it is permissible to multiply things out in the usual manner to arrive at the result:
- Ef(x, y, dx, dy) = x y + x dy + y dx + dx dy
To understand what this means in logical terms, for instance, as expressed in a boolean expansion or a disjunctive normal form (DNF), it is perhaps a little better to go back and analyze the expression the same way that we did for Df. Thus, let us compute the value of the enlarged proposition Ef at each of the points in the universe of discourse U = X × Y.
o-------------------------------------------------o | | | x dx y dy | | o---o o---o | | \ | | / | | \ | | / | | \| |/ | | @=@ | | | o-------------------------------------------------o | Ef = (x, dx) (y, dy) | o-------------------------------------------------o
o-------------------------------------------------o | | | dx dy | | o---o o---o | | \ | | / | | \ | | / | | \| |/ | | @=@ | | | o-------------------------------------------------o | Ef|xy = (dx) (dy) | o-------------------------------------------------o
o-------------------------------------------------o | | | o | | dx | dy | | o---o o---o | | \ | | / | | \ | | / | | \| |/ | | @=@ | | | o-------------------------------------------------o | Ef|x(y) = (dx) dy | o-------------------------------------------------o
o-------------------------------------------------o | | | o | | | dx dy | | o---o o---o | | \ | | / | | \ | | / | | \| |/ | | @=@ | | | o-------------------------------------------------o | Ef|(x)y = dx (dy) | o-------------------------------------------------o
o-------------------------------------------------o | | | o o | | | dx | dy | | o---o o---o | | \ | | / | | \ | | / | | \| |/ | | @=@ | | | o-------------------------------------------------o | Ef|(x)(y) = dx dy | o-------------------------------------------------o
Given the sort of data that arises from this form of analysis, we can now fold the disjoined ingredients back into a boolean expansion or a DNF that is equivalent to the proposition Ef.
- Ef = xy Efxy + x(y) Efx(y) + (x)y Ef(x)y + (x)(y) Ef(x)(y)
Here is a summary of the result, illustrated by means of a digraph picture, where the "no change" element (dx)(dy) is drawn as a loop at the point x y.
o-------------------------------------------------o | f = x y | o-------------------------------------------------o | | | Ef = x y (dx)(dy) | | | | + x (y) (dx) dy | | | | + (x) y dx (dy) | | | | + (x)(y) dx dy | | | o-------------------------------------------------o | | | (dx) (dy) | | .--->---. | | \ / | | \x y/ | | \ / | | x (y) o-------------->o<--------------o (x) y | | (dx) dy ^ dx (dy) | | | | | | | | dx | dy | | | | | | | | | | | o | | (x) (y) | | | o-------------------------------------------------o
We may understand the enlarged proposition Ef as telling us all the different ways to reach a model of f from any point of the universe U.
To broaden our experience with simple examples, let us now contemplate the sixteen functions of concrete type X × Y → B and abstract type B × B → B. For future reference, I will set here a few tables that detail the actions of E and D and on each of these functions, allowing us to view the results in several different ways.
By way of initial orientation, Table 1 lists equivalent expressions for the sixteen functions in a number of different languages for zeroth order logic.
| L1 | L2 | L3 | L4 | L5 | L6 |
|---|---|---|---|---|---|
| x : | 1 1 0 0 | ||||
| y : | 1 0 1 0 | ||||
| f0 | f0000 | 0 0 0 0 | ( ) | false | 0 |
| f1 | f0001 | 0 0 0 1 | (x)(y) | neither x nor y | ¬x ∧ ¬y |
| f2 | f0010 | 0 0 1 0 | (x) y | y and not x | ¬x ∧ y |
| f3 | f0011 | 0 0 1 1 | (x) | not x | ¬x |
| f4 | f0100 | 0 1 0 0 | x (y) | x and not y | x ∧ ¬y |
| f5 | f0101 | 0 1 0 1 | (y) | not y | ¬y |
| f6 | f0110 | 0 1 1 0 | (x, y) | x not equal to y | x ≠ y |
| f7 | f0111 | 0 1 1 1 | (x y) | not both x and y | ¬x ∨ ¬y |
| f8 | f1000 | 1 0 0 0 | x y | x and y | x ∧ y |
| f9 | f1001 | 1 0 0 1 | ((x, y)) | x equal to y | x = y |
| f10 | f1010 | 1 0 1 0 | y | y | y |
| f11 | f1011 | 1 0 1 1 | (x (y)) | not x without y | x → y |
| f12 | f1100 | 1 1 0 0 | x | x | x |
| f13 | f1101 | 1 1 0 1 | ((x) y) | not y without x | x ← y |
| f14 | f1110 | 1 1 1 0 | ((x)(y)) | x or y | x ∨ y |
| f15 | f1111 | 1 1 1 1 | (( )) | true | 1 |
The next four Tables expand the expressions of Ef and Df in two different ways, for each of the sixteen functions. Notice that the functions are given in a different order, here being collected into a set of seven natural classes.
Table 2. Ef Expanded Over Ordinary Features {x, y}
o------o------------o------------o------------o------------o------------o
| | | | | | |
| | f | Ef | xy | Ef | x(y) | Ef | (x)y | Ef | (x)(y)|
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_0 | () | () | () | () | () |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_1 | (x)(y) | dx dy | dx (dy) | (dx) dy | (dx)(dy) |
| | | | | | |
| f_2 | (x) y | dx (dy) | dx dy | (dx)(dy) | (dx) dy |
| | | | | | |
| f_4 | x (y) | (dx) dy | (dx)(dy) | dx dy | dx (dy) |
| | | | | | |
| f_8 | x y | (dx)(dy) | (dx) dy | dx (dy) | dx dy |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_3 | (x) | dx | dx | (dx) | (dx) |
| | | | | | |
| f_12 | x | (dx) | (dx) | dx | dx |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_6 | (x, y) | (dx, dy) | ((dx, dy)) | ((dx, dy)) | (dx, dy) |
| | | | | | |
| f_9 | ((x, y)) | ((dx, dy)) | (dx, dy) | (dx, dy) | ((dx, dy)) |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_5 | (y) | dy | (dy) | dy | (dy) |
| | | | | | |
| f_10 | y | (dy) | dy | (dy) | dy |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_7 | (x y) | ((dx)(dy)) | ((dx) dy) | (dx (dy)) | (dx dy) |
| | | | | | |
| f_11 | (x (y)) | ((dx) dy) | ((dx)(dy)) | (dx dy) | (dx (dy)) |
| | | | | | |
| f_13 | ((x) y) | (dx (dy)) | (dx dy) | ((dx)(dy)) | ((dx) dy) |
| | | | | | |
| f_14 | ((x)(y)) | (dx dy) | (dx (dy)) | ((dx) dy) | ((dx)(dy)) |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_15 | (()) | (()) | (()) | (()) | (()) |
| | | | | | |
o------o------------o------------o------------o------------o------------o
Table 3. Df Expanded Over Ordinary Features {x, y}
o------o------------o------------o------------o------------o------------o
| | | | | | |
| | f | Df | xy | Df | x(y) | Df | (x)y | Df | (x)(y)|
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_0 | () | () | () | () | () |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_1 | (x)(y) | dx dy | dx (dy) | (dx) dy | ((dx)(dy)) |
| | | | | | |
| f_2 | (x) y | dx (dy) | dx dy | ((dx)(dy)) | (dx) dy |
| | | | | | |
| f_4 | x (y) | (dx) dy | ((dx)(dy)) | dx dy | dx (dy) |
| | | | | | |
| f_8 | x y | ((dx)(dy)) | (dx) dy | dx (dy) | dx dy |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_3 | (x) | dx | dx | dx | dx |
| | | | | | |
| f_12 | x | dx | dx | dx | dx |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_6 | (x, y) | (dx, dy) | (dx, dy) | (dx, dy) | (dx, dy) |
| | | | | | |
| f_9 | ((x, y)) | (dx, dy) | (dx, dy) | (dx, dy) | (dx, dy) |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_5 | (y) | dy | dy | dy | dy |
| | | | | | |
| f_10 | y | dy | dy | dy | dy |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_7 | (x y) | ((dx)(dy)) | (dx) dy | dx (dy) | dx dy |
| | | | | | |
| f_11 | (x (y)) | (dx) dy | ((dx)(dy)) | dx dy | dx (dy) |
| | | | | | |
| f_13 | ((x) y) | dx (dy) | dx dy | ((dx)(dy)) | (dx) dy |
| | | | | | |
| f_14 | ((x)(y)) | dx dy | dx (dy) | (dx) dy | ((dx)(dy)) |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_15 | (()) | () | () | () | () |
| | | | | | |
o------o------------o------------o------------o------------o------------o
Table 4. Ef Expanded Over Differential Features {dx, dy}
o------o------------o------------o------------o------------o------------o
| | | | | | |
| | f | T_11 f | T_10 f | T_01 f | T_00 f |
| | | | | | |
| | | Ef| dx dy | Ef| dx(dy) | Ef| (dx)dy | Ef|(dx)(dy)|
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_0 | () | () | () | () | () |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_1 | (x)(y) | x y | x (y) | (x) y | (x)(y) |
| | | | | | |
| f_2 | (x) y | x (y) | x y | (x)(y) | (x) y |
| | | | | | |
| f_4 | x (y) | (x) y | (x)(y) | x y | x (y) |
| | | | | | |
| f_8 | x y | (x)(y) | (x) y | x (y) | x y |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_3 | (x) | x | x | (x) | (x) |
| | | | | | |
| f_12 | x | (x) | (x) | x | x |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_6 | (x, y) | (x, y) | ((x, y)) | ((x, y)) | (x, y) |
| | | | | | |
| f_9 | ((x, y)) | ((x, y)) | (x, y) | (x, y) | ((x, y)) |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_5 | (y) | y | (y) | y | (y) |
| | | | | | |
| f_10 | y | (y) | y | (y) | y |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_7 | (x y) | ((x)(y)) | ((x) y) | (x (y)) | (x y) |
| | | | | | |
| f_11 | (x (y)) | ((x) y) | ((x)(y)) | (x y) | (x (y)) |
| | | | | | |
| f_13 | ((x) y) | (x (y)) | (x y) | ((x)(y)) | ((x) y) |
| | | | | | |
| f_14 | ((x)(y)) | (x y) | (x (y)) | ((x) y) | ((x)(y)) |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_15 | (()) | (()) | (()) | (()) | (()) |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | |
| Fixed Point Total | 4 | 4 | 4 | 16 |
| | | | | |
o-------------------o------------o------------o------------o------------o
Table 5. Df Expanded Over Differential Features {dx, dy}
o------o------------o------------o------------o------------o------------o
| | | | | | |
| | f | Df| dx dy | Df| dx(dy) | Df| (dx)dy | Df|(dx)(dy)|
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_0 | () | () | () | () | () |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_1 | (x)(y) | ((x, y)) | (y) | (x) | () |
| | | | | | |
| f_2 | (x) y | (x, y) | y | (x) | () |
| | | | | | |
| f_4 | x (y) | (x, y) | (y) | x | () |
| | | | | | |
| f_8 | x y | ((x, y)) | y | x | () |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_3 | (x) | (()) | (()) | () | () |
| | | | | | |
| f_12 | x | (()) | (()) | () | () |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_6 | (x, y) | () | (()) | (()) | () |
| | | | | | |
| f_9 | ((x, y)) | () | (()) | (()) | () |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_5 | (y) | (()) | () | (()) | () |
| | | | | | |
| f_10 | y | (()) | () | (()) | () |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_7 | (x y) | ((x, y)) | y | x | () |
| | | | | | |
| f_11 | (x (y)) | (x, y) | (y) | x | () |
| | | | | | |
| f_13 | ((x) y) | (x, y) | y | (x) | () |
| | | | | | |
| f_14 | ((x)(y)) | ((x, y)) | (y) | (x) | () |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_15 | (()) | () | () | () | () |
| | | | | | |
o------o------------o------------o------------o------------o------------o
If the medium truly is the message, the blank slate is the innate idea.
If you think that I linger in the realm of logical difference calculus out of sheer vacillation about getting down to the differential proper, it is probably out of a prior expectation that you derive from the art or the long-engrained practice of real analysis. But the fact is that ordinary calculus only rushes on to the sundry orders of approximation because the strain of comprehending the full import of E and D at once whelm over its discrete and finite powers to grasp them. But here, in the fully serene idylls of ZOL, we find ourselves fit with the compass of a wit that is all we'd ever wish to explore their effects with care.
So let us do just that.
I will first rationalize the novel grouping of propositional forms in the last set of Tables, as that will extend a gentle invitation to the mathematical subject of group theory, and demonstrate its relevance to differential logic in a strikingly apt and useful way. The data for that account is contained in Table 4.
Table 4. Ef Expanded Over Differential Features {dx, dy}
o------o------------o------------o------------o------------o------------o
| | | | | | |
| | f | T_11 f | T_10 f | T_01 f | T_00 f |
| | | | | | |
| | | Ef| dx dy | Ef| dx(dy) | Ef| (dx)dy | Ef|(dx)(dy)|
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_0 | () | () | () | () | () |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_1 | (x)(y) | x y | x (y) | (x) y | (x)(y) |
| | | | | | |
| f_2 | (x) y | x (y) | x y | (x)(y) | (x) y |
| | | | | | |
| f_4 | x (y) | (x) y | (x)(y) | x y | x (y) |
| | | | | | |
| f_8 | x y | (x)(y) | (x) y | x (y) | x y |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_3 | (x) | x | x | (x) | (x) |
| | | | | | |
| f_12 | x | (x) | (x) | x | x |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_6 | (x, y) | (x, y) | ((x, y)) | ((x, y)) | (x, y) |
| | | | | | |
| f_9 | ((x, y)) | ((x, y)) | (x, y) | (x, y) | ((x, y)) |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_5 | (y) | y | (y) | y | (y) |
| | | | | | |
| f_10 | y | (y) | y | (y) | y |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_7 | (x y) | ((x)(y)) | ((x) y) | (x (y)) | (x y) |
| | | | | | |
| f_11 | (x (y)) | ((x) y) | ((x)(y)) | (x y) | (x (y)) |
| | | | | | |
| f_13 | ((x) y) | (x (y)) | (x y) | ((x)(y)) | ((x) y) |
| | | | | | |
| f_14 | ((x)(y)) | (x y) | (x (y)) | ((x) y) | ((x)(y)) |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | | |
| f_15 | (()) | (()) | (()) | (()) | (()) |
| | | | | | |
o------o------------o------------o------------o------------o------------o
| | | | | |
| Fixed Point Total | 4 | 4 | 4 | 16 |
| | | | | |
o-------------------o------------o------------o------------o------------o
The shift operator E can be understood as enacting a substitution operation on the proposition that is given as its argument. In our immediate example, we have the following data and definition:
- E : (U → B) → (EU → B),
- E : f(x, y) → Ef(x, y, dx, dy),
- Ef(x, y, dx, dy) = f(x + dx, y + dy).
Therefore, if we evaluate Ef at particular values of dx and dy, for example, dx = i and dy = j, where i, j are in B, we obtain:
- Eij : (U → B) → (U → B),
- Eij : f → : Eijf,
- Eijf = Ef|<dx = i, dy = j> = f(x + i, y + j).
The notation is a little bit awkward, but the data of the Table should make the sense clear. The important thing to observe is that Eij has the effect of transforming each proposition f : U → B into some other proposition f´ : U → B. As it happens, the action is one-to-one and onto for each Eij, so the gang of four operators {Eij : i, j in B} is an example of what is called a transformation group on the set of sixteen propositions. Bowing to a longstanding local and linear tradition, I will therefore redub the four elements of this group as T00, T01, T10, T11, to bear in mind their transformative character, or nature, as the case may be. Abstractly viewed, this group of order four has the following operation table:
o----------o----------o----------o----------o----------o | % | | | | | * % T_00 | T_01 | T_10 | T_11 | | % | | | | o==========o==========o==========o==========o==========o | % | | | | | T_00 % T_00 | T_01 | T_10 | T_11 | | % | | | | o----------o----------o----------o----------o----------o | % | | | | | T_01 % T_01 | T_00 | T_11 | T_10 | | % | | | | o----------o----------o----------o----------o----------o | % | | | | | T_10 % T_10 | T_11 | T_00 | T_01 | | % | | | | o----------o----------o----------o----------o----------o | % | | | | | T_11 % T_11 | T_10 | T_01 | T_00 | | % | | | | o----------o----------o----------o----------o----------o
It happens that there are just two possible groups of 4 elements. One is the cyclic group Z4 (German Zyklus), which this is not. The other is Klein's four-group V4 (German Vier), which it is.
More concretely viewed, the group as a whole pushes the set of sixteen propositions around in such a way that they fall into seven natural classes, called orbits. One says that the orbits are preserved by the action of the group. There is an Orbit Lemma of immense utility to "those who count" which, depending on your upbringing, you may associate with the names of Burnside, Cauchy, Frobenius, or some subset or superset of these three, vouching that the number of orbits is equal to the mean number of fixed points, in other words, the total number of points (in our case, propositions) that are left unmoved by the separate operations, divided by the order of the group. In this instance, T00 operates as the group identity, fixing all 16 propositions, while the other three group elements fix 4 propositions each, and so we get:
- Number of orbits = (4 + 4 + 4 + 16) ÷ 4 = 7.
Amazing!
We have been contemplating functions of the type f : U → B, studying the action of the operators E and D on this family. These functions, that we may identify for our present aims with propositions, inasmuch as they capture their abstract forms, are logical analogues of scalar potential fields. These are the sorts of fields that are so picturesquely presented in elementary calculus and physics textbooks by images of snow-covered hills and parties of skiers who trek down their slopes like least action heroes. The analogous scene in propositional logic presents us with forms more reminiscent of plateaunic idylls, being all plains at one of two levels, the mesas of verity and falsity, as it were, with nary a niche to inhabit between them, restricting our options for a sporting gradient of downhill dynamics to just one of two, standing still on level ground or falling off a bluff.
We are still working well within the logical analogue of the classical finite difference calculus, taking in the novelties that the logical transmutation of familiar elements is able to bring to light. Soon we will take up several different notions of approximation relationships that may be seen to organize the space of propositions, and these will allow us to define several different forms of differential analysis applying to propositions. In time we will find reason to consider more general types of maps, having concrete types of the form X1 × … × Xk → Y1 × … × Yn and abstract types Bk → Bn. We will think of these mappings as transforming universes of discourse into themselves or into others, in short, as transformations of discourse.
Before we continue with this intinerary, however, I would like to highlight another sort of differential aspect that concerns the boundary operator or the marked connective that serves as one of the two basic connectives in the cactus language for ZOL.
For example, consider the proposition f of concrete type f : X × Y × Z → B and abstract type f : B3 → B that is written (x, y, z) in cactus syntax. Taken as an assertion in what Peirce called the existential interpretation, (x, y, z) says that just one of x, y, z is false. It is useful to consider this assertion in relation to the conjunction xyz of the features that are engaged as its arguments. A venn diagram of (x, y, z) looks like this:
o-----------------------------------------------------------o | U | | | | o-------------o | | / \ | | / \ | | / \ | | / \ | | / \ | | o x o | | | | | | | | | | | | | | | | | | | | | | o--o----------o o----------o--o | | / \%%%%%%%%%%\ /%%%%%%%%%%/ \ | | / \%%%%%%%%%%o%%%%%%%%%%/ \ | | / \%%%%%%%%/ \%%%%%%%%/ \ | | / \%%%%%%/ \%%%%%%/ \ | | / \%%%%/ \%%%%/ \ | | o o--o-------o--o o | | | |%%%%%%%| | | | | |%%%%%%%| | | | | |%%%%%%%| | | | | |%%%%%%%| | | | | |%%%%%%%| | | | o y o%%%%%%%o z o | | \ \%%%%%/ / | | \ \%%%/ / | | \ \%/ / | | \ o / | | \ / \ / | | o-------------o o-------------o | | | | | o-----------------------------------------------------------o
In relation to the center cell indicated by the conjunction xyz, the region indicated by (x, y, z) is comprised of the adjacent or the bordering cells. Thus they are the cells that are just across the boundary of the center cell, as if reached by way of Leibniz's minimal changes from the point of origin, here, xyz.
The same form of boundary relationship is exhibited for any cell of origin that one might elect to indicate, say, by means of the conjunction of positive and negative basis features u1 … uk, where uj = xj or uj = (xj), for j = 1 to k. The proposition (u1, …, uk) indicates the disjunctive region consisting of the cells that are "just next door" to the cell u1 … uk.
The Pragmatic Maxim
Consider what effects that might conceivably have practical bearings you conceive the objects of your conception to have. Then, your conception of those effects is the whole of your conception of the object.
Charles Sanders Peirce, "The Maxim of Pragmatism, CP 5.438.
One other subject that it would be opportune to mention at this point, while we have an object example of a mathematical group fresh in mind, is the relationship between the pragmatic maxim and what are commonly known in mathematics as representation principles. As it turns out, with regard to its formal characteristics, the pragmatic maxim unites the aspects of a representation principle with the attributes of what would ordinarily be known as a closure principle. We will consider the form of closure that is invoked by the pragmatic maxim on another occasion, focusing here and now on the topic of group representations.
Let us return to the example of the so-called four-group V4. We encountered this group in one of its concrete representations, namely, as a transformation group that acts on a set of objects, in this particular case a set of sixteen functions or propositions. Forgetting about the set of objects that the group transforms among themselves, we may take the abstract view of the group's operational structure, say, in the form of the group operation table copied here:
o---------o---------o---------o---------o---------o | % | | | | | . % e | f | g | h | | % | | | | o=========o=========o=========o=========o=========o | % | | | | | e % e | f | g | h | | % | | | | o---------o---------o---------o---------o---------o | % | | | | | f % f | e | h | g | | % | | | | o---------o---------o---------o---------o---------o | % | | | | | g % g | h | e | f | | % | | | | o---------o---------o---------o---------o---------o | % | | | | | h % h | g | f | e | | % | | | | o---------o---------o---------o---------o---------o
This table is abstractly the same as, or isomorphic to, the versions with the Eij operators and the Tij transformations that we discussed earlier. That is to say, the story is the same — only the names have been changed. An abstract group can have a multitude of significantly and superficially different representations. Even after we have long forgotten the details of the particular representation that we may have come in with, there are species of concrete representations, called the regular representations, that are always readily available, as they can be generated from the mere data of the abstract operation table itself.
For example, select a group element from the top margin of the Table, and "consider its effects" on each of the group elements as they are listed along the left margin. We may record these effects as Peirce usually did, as a logical "aggregate" of elementary dyadic relatives, that is to say, a disjunction or a logical sum whose terms represent the ordered pairs of <input : output> transactions that are produced by each group element in turn. This yields what is usually known as one of the regular representations of the group, specifically, the first, the post-, or the right regular representation. It has long been conventional to organize the terms in the form of a matrix:
Reading "+" as a logical disjunction:
G = e + f + g + h,
And so, by expanding effects, we get:
G = e:e + f:f + g:g + h:h
+ e:f + f:e + g:h + h:g
+ e:g + f:h + g:e + h:f
+ e:h + f:g + g:f + h:e
More on the pragmatic maxim as a representation principle later.
Consider what effects that might conceivably have practical bearings you conceive the objects of your conception to have. Then, your conception of those effects is the whole of your conception of the object.
Peirce, "Maxim of Pragmaticism", Collected Papers, CP 5.438.
The genealogy of this conception of pragmatic representation is very intricate. I will delineate some details that I presently fancy I remember clearly enough, subject to later correction. Without checking historical accounts, I will not be able to pin down anything like a real chronology, but most of these notions were standard furnishings of the 19th Century mathematical study, and only the last few items date as late as the 1920's.
The idea about the regular representations of a group is universally known as Cayley's Theorem, usually in the form: "Every group is isomorphic to a subgroup of Aut(X), the group of automorphisms of an appropriate set X". There is a considerable generalization of these regular representations to a broad class of relational algebraic systems in Peirce's earliest papers. The crux of the whole idea is this:
Contemplate the effects of the symbol whose meaning you wish to investigate as they play out on all the stages of conduct on which you have the ability to imagine that symbol playing a role.
This idea of contextual definition is basically the same as Jeremy Bentham's notion of paraphrasis, a "method of accounting for fictions by explaining various purported terms away" (Quine, in Van Heijenoort, p. 216). Today we'd call these constructions term models. This, again, is the big idea behind Schönfinkel's combinators {S, K, I}, and hence of lambda calculus, and I reckon you know where that leads.
Let me return to Peirce's early papers on the algebra of relatives to pick up the conventions that he used there, and then rewrite my account of regular representations in a way that conforms to those.
Peirce expresses the action of an "elementary dual relative" like so:
[Let] A:B be taken to denote the elementary relative which multiplied into B gives A. (Peirce, CP 3.123).
And though he is well aware that it is not at all necessary to arrange elementary relatives into arrays, matrices, or tables, when he does so he tends to prefer organizing dyadic relations in the following manner:
[ A:A A:B A:C | | | | B:A B:B B:C | | | | C:A C:B C:C ]
That conforms to the way that the last school of thought I matriculated into stipulated that we tabulate material:
[ e_11 e_12 e_13 | | | | e_21 e_22 e_23 | | | | e_31 e_32 e_33 ]
So, for example, let us suppose that we have the small universe {A, B, C}, and the 2-adic relation m = mover of that is represented by this matrix:
m = [ m_AA (A:A) m_AB (A:B) m_AC (A:C) | | | | m_BA (B:A) m_BB (B:B) m_BC (B:C) | | | | m_CA (C:A) m_CB (C:B) m_CC (C:C) ]
Also, let m be such that:
A is a mover of A and B, B is a mover of B and C, C is a mover of C and A.
In sum:
m = [ 1 * (A:A) 1 * (A:B) 0 * (A:C) | | | | 0 * (B:A) 1 * (B:B) 1 * (B:C) | | | | 1 * (C:A) 0 * (C:B) 1 * (C:C) ]
For the sake of orientation and motivation, compare with Peirce's notation in CP 3.329.
I think that will serve to fix notation and set up the remainder of the account.
It is common in algebra to switch around between different conventions of display, as the momentary fancy happens to strike, and I see that Peirce is no different in this sort of shiftiness than anyone else. A changeover appears to occur especially whenever he shifts from logical contexts to algebraic contexts of application.
In the paper "On the Relative Forms of Quaternions" (CP 3.323), we observe Peirce providing the following sorts of explanation:
If X, Y, Z denote the three rectangular components of a vector, and W' denote numerical unity (or a fourth rectangular component, involving space of four dimensions), and (Y:Z) denote the operation of converting the Y component of a vector into its Z component, then
1 = (W:W) + (X:X) + (Y:Y) + (Z:Z) i = (X:W) - (W:X) - (Y:Z) + (Z:Y) j = (Y:W) - (W:Y) - (Z:X) + (X:Z) k = (Z:W) - (W:Z) - (X:Y) + (Y:X)In the language of logic (Y:Z) is a relative term whose relate is a Y component, and whose correlate is a Z component. The law of multiplication is plainly (Y:Z)(Z:X) = (Y:X), (Y:Z)(X:W) = 0, and the application of these rules to the above values of 1, i, j, k gives the quaternion relations
i^2 = j^2 = k^2 = -1, ijk = -1, etc.The symbol a(Y:Z) denotes the changing of Y to Z and the multiplication of the result by a'. If the relatives be arranged in a block
W:W W:X W:Y W:Z X:W X:X X:Y X:Z Y:W Y:X Y:Y Y:Z Z:W Z:X Z:Y Z:Zthen the quaternion w + xi + yj + zk is represented by the matrix of numbers
w -x -y -z x w -z y y z w -x z -y x wThe multiplication of such matrices follows the same laws as the multiplication of quaternions. The determinant of the matrix = the fourth power of the tensor of the quaternion.
The imaginary x + y(-1)^(1/2) may likewise be represented by the matrix
x y -y xand the determinant of the matrix = the square of the modulus.
C.S. Peirce, Collected Papers, CP 3.323, (1882). Johns Hopkins University Circulars, No. 13, p. 179.
This way of talking is the mark of a person who opts to multiply his matrices "on the right", as they say. Yet Peirce still continues to call the first element of the ordered pair (i:j) its "relate" while calling the second element of the pair (i:j) its "correlate". That doesn't comport very well, so far as I can tell, with his customary reading of relative terms, suited more to the multiplication of matrices "on the left".
So I still have a few wrinkles to iron out before I can give this story a smooth enough consistency.
Let us make up the model universe $1$ = A + B + C and the 2-adic relation n = "noter of", as when "X is a data record that contains a pointer to Y". That interpretation is not important, it's just for the sake of intuition. In general terms, the 2-adic relation n can be represented by this matrix:
n = [ n_AA (A:A) n_AB (A:B) n_AC (A:C) | | | | n_BA (B:A) n_BB (B:B) n_BC (B:C) | | | | n_CA (C:A) n_CB (C:B) n_CC (C:C) ]
Also, let n be such that:
A is a noter of A and B, B is a noter of B and C, C is a noter of C and A.
Filling in the instantial values of the "coefficients" nij, as the indices i and j range over the universe of discourse:
n = [ 1 * (A:A) 1 * (A:B) 0 * (A:C) | | | | 0 * (B:A) 1 * (B:B) 1 * (B:C) | | | | 1 * (C:A) 0 * (C:B) 1 * (C:C) ]
In Peirce's time, and even in some circles of mathematics today, the information indicated by the elementary relatives (i:j), as i, j range over the universe of discourse, would be referred to as the "umbral elements" of the algebraic operation represented by the matrix, though I seem to recall that Peirce preferred to call these terms the "ingredients". When this ordered basis is understood well enough, one will tend to drop any mention of it from the matrix itself, leaving us nothing but these bare bones:
n = [ 1 1 0 | | | | 0 1 1 | | | | 1 0 1 ]
However the specification may come to be written, this is all just convenient schematics for stipulating that:
- n = A:A + B:B + C:C'+ A:B + B:C + C:A
Recognizing !1! = A:A + B:B + C:C to be the identity transformation, the 2-adic relation n = "noter of" may be represented by an element !1! + A:B + B:C + C:A of the so-called "group ring", all of which just makes this element a special sort of linear transformation.
Up to this point, we are still reading the elementary relatives of the form i:j in the way that Peirce reads them in logical contexts: i is the relate, j is the correlate, and in our current example we read i:j, or more exactly, nij = 1, to say that i is a noter of j. This is the mode of reading that we call "multiplying on the left".
In the algebraic, permutational, or transformational contexts of application, however, Peirce converts to the alternative mode of reading, although still calling i the relate and j the correlate, the elementary relative i:j now means that i gets changed into j. In this scheme of reading, the transformation A:B + B:C + C:A is a permutation of the aggregate $1$ = A + B + C, or what we would now call the set {A, B, C}, in particular, it is the permutation that is otherwise notated as:
( A B C ) < > ( B C A )
This is consistent with the convention that Peirce uses in the paper "On a Class of Multiple Algebras" (CP 3.324–327).
We have been contemplating the virtues and the utilities of the pragmatic maxim as a standard heuristic in hermeneutics, that is, as a principle of interpretation that guides us in finding clarifying representations for a problematic corpus of symbols by means of their actions on other symbols or in terms of their effects on the syntactic contexts wherein we discover them or where we might conceive to distribute them.
I began this excursion by taking off from the moving platform of differential logic and passing by way of the corresponding transformation groups, as they act on propositions, and on to an exercise in applying the pragmatic maxim, by contemplating the regular representations of groups as giving us one of the simplest conceivable, relatively concrete applications of the general principle of representation in question.
There are a few problems of implementation that have to be worked out in practice, most of which are cleared up by keeping in mind which of several possible conventions we have chosen to follow at a given time.
But there does appear to remain this rather more substantial question: Are the effects we seek relates or correlates, or does it even matter?
I will have to leave that question as it is for now, in hopes that a solution will evolve itself in time.
Obstacles to Applying the Pragmatic Maxim
No sooner do you get a good idea and try to apply it than you find that a motley array of obstacles arise.
It would be good if we could in practice more consistently apply the pragmatic maxim to the purpose for which it was purportedly intended by its author. That aim would be the clarification of concepts, that is, intellectual symbols or mental signs, to the point where their inherent senses, or their lacks thereof, would be rendered manifest to suitable interpreters.
There are big obstacles and little obstacles to applying the pragmatic maxim. In good subgoaling fashion, I will merely mention a few of the bigger blocks, as if in passing, but not really getting past them, and then I will get down to the details of the problems that more immediately obstruct our advance.
Obstacle 1. People do not always read the instructions very carefully. There is a tendency in readers of particular prior persuasions to blow the problem all out of proportion, to think that the maxim is meant to reveal the absolutely positive and the totally unique meaning of every preconception to which they might deign or elect to apply it. Reading the maxim with an even minimal attention, you can see that it promises no such finality of unindexed sense, but ties what you conceive to you. I have lately come to wonder at the tenacity of this misinterpretation. Perhaps people reckon that nothing less would be worth their attention. I am not sure. I can only say the achievement of more modest goals is the sort of thing on which our daily life depends, and there can be no final end to inquiry nor any ultimate community without a continuation of life, and that means life on a day to day basis. All of which only brings me back to the point of persisting with local meantime examples, because if we can't apply the maxim there, we can't apply it anywhere.
Obstacle 2. Applying the pragmatic maxim, even with a moderate aim, can be hard. I think that my present example, deliberately impoverished as it is, affords us with an embarassing richness of evidence of just how complex the simple can be.
All the better reason for me to see if I can finish it up before moving on.
Expressed most simply, the idea is to replace the question of "what it is", which modest people know is far too difficult for them to answer right off, with the question of "what it does", which most of us know a modicum about.
In the case of regular representations of groups we found a non-plussing surplus of answers to sort our way through. So let us track back one more time to see if we can learn any lessons that might carry over to more realistic cases.
Here is is the operation table of V4 once again:
Table 1. Klein Four-Group V_4 o---------o---------o---------o---------o---------o | % | | | | | . % e | f | g | h | | % | | | | o=========o=========o=========o=========o=========o | % | | | | | e % e | f | g | h | | % | | | | o---------o---------o---------o---------o---------o | % | | | | | f % f | e | h | g | | % | | | | o---------o---------o---------o---------o---------o | % | | | | | g % g | h | e | f | | % | | | | o---------o---------o---------o---------o---------o | % | | | | | h % h | g | f | e | | % | | | | o---------o---------o---------o---------o---------o
A group operation table is really just a device for recording a certain 3-adic relation, to be specific, the set of triples of the form (x, y, z) satisfying the equation x\(\cdot\)y = z, where "\(\cdot\)" signifies the group operation, usually omitted as understood in context.
In the case of V4 = (G, \(\cdot\)), where G is the "underlying set" {e, f, g, h}, we have the 3-adic relation L(V4) ⊆ G × G × G whose triples are listed below:
<e, e, e> <e, f, f> <e, g, g> <e, h, h> <f, e, f> <f, f, e> <f, g, h> <f, h, g> <g, e, g> <g, f, h> <g, g, e> <g, h, f> <h, e, h> <h, f, g> <h, g, f> <h, h, e>
It is part of the definition of a group that the 3-adic relation L ⊆ G3 is actually a function L : G × G → G. It is from this functional perspective that we can see an easy way to derive the two regular representations. Since we have a function of the type L : G × G → G, we can define a couple of substitution operators:
- Sub(x, (_, y)) puts any specified x into the empty slot of the rheme (_, y), with the effect of producing the saturated rheme (x, y) that evaluates to xy.
- Sub(x, (y, _)) puts any specified x into the empty slot of the rheme (y, _), with the effect of producing the saturated rheme (y, x) that evaluates to yx.
In (1), we consider the effects of each x in its practical bearing on contexts of the form (_, y), as y ranges over G, and the effects are such that x takes (_, y) into xy, for y in G, all of which is summarily notated as x = {(y : xy) : y in G}. The pairs (y : xy) can be found by picking an x from the left margin of the group operation table and considering its effects on each y in turn as these run across the top margin. This aspect of pragmatic definition we recognize as the regular ante-representation:
e = e:e + f:f + g:g + h:h f = e:f + f:e + g:h + h:g g = e:g + f:h + g:e + h:f h = e:h + f:g + g:f + h:e
In (2), we consider the effects of each x in its practical bearing on contexts of the form (y, _), as y ranges over G, and the effects are such that x takes (y, _) into yx, for y in G, all of which is summarily notated as x = {(y : yx) : y in G}. The pairs (y : yx) can be found by picking an x from the top margin of the group operation table and considering its effects on each y in turn as these run down the left margin. This aspect of pragmatic definition we recognize as the regular post-representation:
e = e:e + f:f + g:g + h:h f = e:f + f:e + g:h + h:g g = e:g + f:h + g:e + h:f h = e:h + f:g + g:f + h:e
If the ante-rep looks the same as the post-rep, now that I'm writing them in the same dialect, that is because V4 is abelian (commutative), and so the two representations have the very same effects on each point of their bearing.
So long as we're in the neighborhood, we might as well take in some more of the sights, for instance, the smallest example of a non-abelian (non-commutative) group. This is a group of six elements, say, G = {e, f, g, h, i, j}, with no relation to any other employment of these six symbols being implied, of course, and it can be most easily represented as the permutation group on a set of three letters, say, X = {A, B, C}, usually notated as G = Sym(X) or more abstractly and briefly, as Sym(3) or S3. Here are the permutation (= substitution) operations in Sym(X):
Table 1. Permutations or Substitutions in Sym_{A, B, C}
o---------o---------o---------o---------o---------o---------o
| | | | | | |
| e | f | g | h | i | j |
| | | | | | |
o=========o=========o=========o=========o=========o=========o
| | | | | | |
| A B C | A B C | A B C | A B C | A B C | A B C |
| | | | | | |
| | | | | | | | | | | | | | | | | | | | | | | | |
| v v v | v v v | v v v | v v v | v v v | v v v |
| | | | | | |
| A B C | C A B | B C A | A C B | C B A | B A C |
| | | | | | |
o---------o---------o---------o---------o---------o---------o
Here is the operation table for S3, given in abstract fashion:
Table 2. Symmetric Group S_3 | ^ | e / \ e | / \ | / e \ | f / \ / \ f | / \ / \ | / f \ f \ | g / \ / \ / \ g | / \ / \ / \ | / g \ g \ g \ | h / \ / \ / \ / \ h | / \ / \ / \ / \ | / h \ e \ e \ h \ | i / \ / \ / \ / \ / \ i | / \ / \ / \ / \ / \ | / i \ i \ f \ j \ i \ | j / \ / \ / \ / \ / \ / \ j | / \ / \ / \ / \ / \ / \ | ( j \ j \ j \ i \ h \ j ) | \ / \ / \ / \ / \ / \ / | \ / \ / \ / \ / \ / \ / | \ h \ h \ e \ j \ i / | \ / \ / \ / \ / \ / | \ / \ / \ / \ / \ / | \ i \ g \ f \ h / | \ / \ / \ / \ / | \ / \ / \ / \ / | \ f \ e \ g / | \ / \ / \ / | \ / \ / \ / | \ g \ f / | \ / \ / | \ / \ / | \ e / | \ / | \ / | v
By the way, we will meet with the symmetric group S3 again when we return to take up the study of Peirce's early paper "On a Class of Multiple Algebras" (CP 3.324–327), and also his late unpublished work "The Simplest Mathematics" (1902) (CP 4.227–323), with particular reference to the section that treats of "Trichotomic Mathematics" (CP 4.307–323).
By way of collecting a short-term pay-off for all the work that we did on the regular representations of the Klein 4-group V4, let us write out as quickly as possible in "relative form" a minimal budget of representations for the symmetric group on three letters, Sym(3). After doing the usual bit of compare and contrast among the various representations, we will have enough concrete material beneath our abstract belts to tackle a few of the presently obscur'd details of Peirce's early "Algebra + Logic" papers.
Table 1. Permutations or Substitutions in Sym {A, B, C}
o---------o---------o---------o---------o---------o---------o
| | | | | | |
| e | f | g | h | i | j |
| | | | | | |
o=========o=========o=========o=========o=========o=========o
| | | | | | |
| A B C | A B C | A B C | A B C | A B C | A B C |
| | | | | | |
| | | | | | | | | | | | | | | | | | | | | | | | |
| v v v | v v v | v v v | v v v | v v v | v v v |
| | | | | | |
| A B C | C A B | B C A | A C B | C B A | B A C |
| | | | | | |
o---------o---------o---------o---------o---------o---------o
Writing this table in relative form generates the following "natural representation" of S3.
e = A:A + B:B + C:C f = A:C + B:A + C:B g = A:B + B:C + C:A h = A:A + B:C + C:B i = A:C + B:B + C:A j = A:B + B:A + C:C
I have without stopping to think about it written out this natural representation of S3 in the style that comes most naturally to me, to wit, the "right" way, whereby an ordered pair configured as X:Y constitutes the turning of X into Y. It is possible that the next time we check in with CSP that we will have to adjust our sense of direction, but that will be an easy enough bridge to cross when we come to it.
To construct the regular representations of S3, we pick up from the data of its operation table:
Table 1. Symmetric Group S_3 | ^ | e / \ e | / \ | / e \ | f / \ / \ f | / \ / \ | / f \ f \ | g / \ / \ / \ g | / \ / \ / \ | / g \ g \ g \ | h / \ / \ / \ / \ h | / \ / \ / \ / \ | / h \ e \ e \ h \ | i / \ / \ / \ / \ / \ i | / \ / \ / \ / \ / \ | / i \ i \ f \ j \ i \ | j / \ / \ / \ / \ / \ / \ j | / \ / \ / \ / \ / \ / \ | ( j \ j \ j \ i \ h \ j ) | \ / \ / \ / \ / \ / \ / | \ / \ / \ / \ / \ / \ / | \ h \ h \ e \ j \ i / | \ / \ / \ / \ / \ / | \ / \ / \ / \ / \ / | \ i \ g \ f \ h / | \ / \ / \ / \ / | \ / \ / \ / \ / | \ f \ e \ g / | \ / \ / \ / | \ / \ / \ / | \ g \ f / | \ / \ / | \ / \ / | \ e / | \ / | \ / | v
Just by way of staying clear about what we are doing, let's return to the recipe that we worked out before:
It is part of the definition of a group that the 3-adic relation L ⊆ G3 is actually a function L : G × G → G. It is from this functional perspective that we can see an easy way to derive the two regular representations.
Since we have a function of the type L : G × G → G, we can define a couple of substitution operators:
- Sub(x, «_, y») puts any specified x into the empty slot of the rheme «_, y», with the effect of producing the saturated rheme «x, y» that evaluates to xy.
- Sub(x, «y, _») puts any specified x into the empty slot of the rheme «y, _», with the effect of producing the saturated rheme «y, x» that evaluates to yx.
In (1), we consider the effects of each x in its practical bearing on contexts of the form «_, y», as y ranges over G, and the effects are such that x takes «_, y» into xy, for y in G, all of which is summarily notated as x = {(y : xy) : y in G}. The pairs (y : xy) can be found by picking an x from the left margin of the group operation table and considering its effects on each y in turn as these run along the right margin. This produces the regular ante-representation of S3, like so:
e = e:e + f:f + g:g + h:h + i:i + j:j f = e:f + f:g + g:e + h:j + i:h + j:i g = e:g + f:e + g:f + h:i + i:j + j:h h = e:h + f:i + g:j + h:e + i:f + j:g i = e:i + f:j + g:h + h:g + i:e + j:f j = e:j + f:h + g:i + h:f + i:g + j:e
In (2), we consider the effects of each x in its practical bearing on contexts of the form «y, _», as y ranges over G, and the effects are such that x takes «y, _» into yx, for y in G, all of which is summarily notated as x = {(y : yx) : y in G}. The pairs (y : yx) can be found by picking an x on the right margin of the group operation table and considering its effects on each y in turn as these run along the left margin. This generates the regular post-representation of S3, like so:
e = e:e + f:f + g:g + h:h + i:i + j:j f = e:f + f:g + g:e + h:i + i:j + j:h g = e:g + f:e + g:f + h:j + i:h + j:i h = e:h + f:j + g:i + h:e + i:g + j:f i = e:i + f:h + g:j + h:f + i:e + j:g j = e:j + f:i + g:h + h:g + i:f + j:e
If the ante-rep looks different from the post-rep, it is just as it should be, as S3 is non-abelian (non-commutative), and so the two representations differ in the details of their practical effects, though, of course, being representations of the same abstract group, they must be isomorphic.
the way of heaven and earth
is to be long continued
in their operation
without stoppingi ching, hexagram 32
You may be wondering what happened to the announced subject of "Differential Logic". If you think that we have been taking a slight excursion my reply to the charge of a scenic rout would be both "yes and no". What happened was this. We chanced to make the observation that the shift operators Eij form a transformation group that acts on the set of propositions of the form f : B2 → B. Group theory is a very attractive subject, but it did not have the effect of drawing us so far off our initial course as one might at first think. For one thing, groups, in particular, the special family of groups that have come to be named after the Norwegian mathematician Marius Sophus Lie, turn out to be of critical importance in the solution of differential equations. For another thing, group operations afford us examples of 3-adic relations that have been extremely well-studied over the years, and thus they supply us with no small bit of guidance in the study of sign relations, another class of 3-adic relations that have significance for logical studies, in our brief acquaintance with which we have scarcely even begun to break the ice. Finally, I could not resist taking up the connection between group representations, which constitute a very generic class of logical models, and the all-important pragmatic maxim.
Biographical Data for Marius Sophus Lie (1842–1899)
We've seen a couple of groups, V4 and S3, represented in various ways, and we've seen their representations presented in a variety of different manners. Let us look at one other stylistic variant for presenting a representation that is frequently seen, the so-called "matrix representation" of a group.
Recalling the manner of our acquaintance with the symmetric group S3, we began with the "bigraph" (bipartite graph) picture of its natural representation as the set of all permutations or substitutions on the set X = {A, B, C}.
Table 1. Permutations or Substitutions in Sym {A, B, C}
o---------o---------o---------o---------o---------o---------o
| | | | | | |
| e | f | g | h | i | j |
| | | | | | |
o=========o=========o=========o=========o=========o=========o
| | | | | | |
| A B C | A B C | A B C | A B C | A B C | A B C |
| | | | | | |
| | | | | | | | | | | | | | | | | | | | | | | | |
| v v v | v v v | v v v | v v v | v v v | v v v |
| | | | | | |
| A B C | C A B | B C A | A C B | C B A | B A C |
| | | | | | |
o---------o---------o---------o---------o---------o---------o
Then we rewrote these permutations — since they are functions f : X → X they can also be recognized as 2-adic relations f ⊆ X × X — in "relative form", in effect, in the manner to which Peirce would have made us accustomed had he been given a relative half-a-chance:
e = A:A + B:B + C:C f = A:C + B:A + C:B g = A:B + B:C + C:A h = A:A + B:C + C:B i = A:C + B:B + C:A j = A:B + B:A + C:C
These days one is much more likely to encounter the natural representation of S3 in the form of a "linear representation", that is, as a family of linear transformations that map the elements of a suitable vector space into each other, all of which would in turn usually be represented by a set of matrices like these:
Table 2. Matrix Representations of the Permutations in Sym(3) o---------o---------o---------o---------o---------o---------o | | | | | | | | e | f | g | h | i | j | | | | | | | | o=========o=========o=========o=========o=========o=========o | | | | | | | | 1 0 0 | 0 0 1 | 0 1 0 | 1 0 0 | 0 0 1 | 0 1 0 | | 0 1 0 | 1 0 0 | 0 0 1 | 0 0 1 | 0 1 0 | 1 0 0 | | 0 0 1 | 0 1 0 | 1 0 0 | 0 1 0 | 1 0 0 | 0 0 1 | | | | | | | | o---------o---------o---------o---------o---------o---------o
The key to the mysteries of these matrices is revealed by noting that their coefficient entries are arrayed and overlayed on a place mat marked like so:
[ A:A A:B A:C | | B:A B:B B:C | | C:A C:B C:C ]
Of course, the place-settings of convenience at different symposia may vary.
Differential Logic : Series B
Note 1
Linear Topics : The Differential Theory of Qualitative Equations
The most fundamental concept in cybernetics is that of "difference", either that two things are recognisably different or that one thing has changed with time.
William Ross Ashby, An Introduction to Cybernetics, Chapman and Hall, London, UK, 1956. Reprinted, Methuen and Company, London, UK, 1964.
This chapter is titled "Linear Topics" because that is the heading under which the derivatives and the differentials of any functions usually come up in mathematics, namely, in relation to the problem of computing "locally linear approximations" to the more arbitrary, unrestricted brands of functions that one finds in a given setting.
To denote lists of propositions and to detail their components, we use notations like:
\[\mathbf{a} = (a, b, c),\ \mathbf{p} = (p, q, r),\ \mathbf{x} = (x, y, z),\!\]
or, in more complicated situations:
\[x = (x_1, x_2, x_3),\ y = (y_1, y_2, y_3),\ z = (z_1, z_2, z_3).\!\]
In a universe where some region is ruled by a proposition, it is natural to ask whether we can change the value of that proposition by changing the features of our current state.
Given a venn diagram with a shaded region and starting from any cell in that universe, what sequences of feature changes, what traverses of cell walls, will take us from shaded to unshaded areas, or the reverse?
In order to discuss questions of this type, it is useful to define several "operators" on functions. An operator is nothing more than a function between sets that happen to have functions as members.
A typical operator \(\operatorname{F}\) takes us from thinking about a given function \(f\!\) to thinking about another function \(g\!\). To express the fact that \(g\!\) can be obtained by applying the operator \(\operatorname{F}\) to \(f\!\), we write \(g = \operatorname{F}f.\)
The first operator, \(\operatorname{E}\), associates with a function \(f : X \to Y\) another function \(\operatorname{E}f\), where \(\operatorname{E}f : X \times X \to Y\) is defined by the following equation:
\[\operatorname{E}f(x, y) = f(x + y).\]
\(\operatorname{E}\) is called a "shift operator" because it takes us from contemplating the value of \(f\!\) at a place \(x\!\) to considering the value of \(f\!\) at a shift of \(y\!\) away. Thus, \(\operatorname{E}\) tells us the absolute effect on \(f\!\) that is obtained by changing its argument from \(x\!\) by an amount that is equal to \(y\!\).
Historical Note. The "shift operator" \(\operatorname{E}\) was originally called the "enlargement operator", hence the initial "E" of the usual notation.
The next operator, \(\operatorname{D}\), associates with a function \(f : X \to Y\) another function \(\operatorname{D}f\), where \(\operatorname{D}f : X \times X \to Y\) is defined by the following equation:
\[\operatorname{D}f(x, y) = \operatorname{E}f(x, y) - f(x),\]
or, equivalently,
\[\operatorname{D}f(x, y) = f(x + y) - f(x).\]
\(\operatorname{D}\) is called a "difference operator" because it tells us about the relative change in the value of \(f\!\) along the shift from \(x\!\) to \(x + y.\!\)
In practice, one of the variables, \(x\!\) or \(y\!\), is often considered to be "less variable" than the other one, being fixed in the context of a concrete discussion. Thus, we might find any one of the following idioms:
\[\operatorname{D}f : X \times X \to Y,\]
\[\operatorname{D}f(c, x) = f(c + x) - f(c).\]
Here, \(c\!\) is held constant and \(\operatorname{D}f(c, x)\) is regarded mainly as a function of the second variable \(x\!\), giving the relative change in \(f\!\) at various distances \(x\!\) from the center \(c\!\).
\[\operatorname{D}f : X \times X \to Y,\]
\[\operatorname{D}f(x, h) = f(x + h) - f(x).\]
Here, \(h\!\) is either a constant (usually 1), in discrete contexts, or a variably "small" amount (near to 0) over which a limit is being taken, as in continuous contexts. \(\operatorname{D}f(x, h)\) is regarded mainly as a function of the first variable \(x\!\), in effect, giving the differences in the value of \(f\!\) between \(x\!\) and a neighbor that is a distance of \(h\!\) away, all the while that \(x\!\) itself ranges over its various possible locations.
\[\operatorname{D}f : X \times X \to Y,\]
\[\operatorname{D}f(x, \operatorname{d}x) = f(x + \operatorname{d}x) - f(x).\]
This is yet another variant of the previous form, with \(\operatorname{d}x\) denoting small changes contemplated in \(x\!\).
That's the basic idea. The next order of business is to develop the logical side of the analogy a bit more fully, and to take up the elaboration of some moderately simple applications of these ideas to a selection of relatively concrete examples.
Note 2
Example 1. A Polymorphous Concept
I start with an example that is simple enough that it will allow us to compare the representations of propositions by venn diagrams, truth tables, and my own favorite version of the syntax for propositional calculus all in a relatively short space. To enliven the exercise, I borrow an example from a book with several independent dimensions of interest, Topobiology by Gerald Edelman. One finds discussed there the notion of a "polymorphous set". Such a set is defined in a universe of discourse whose elements can be described in terms of a fixed number \(k\!\) of logical features. A "polymorphous set" is one that can be defined in terms of sets whose elements have a fixed number \(j\!\) of the \(k\!\) features.
As a rule in the following discussion, I will use upper case letters as names for concepts and sets, lower case letters as names for features and functions.
The example that Edelman gives (1988, Fig. 10.5, p. 194) involves sets of stimulus patterns that can be described in terms of the three features "round" \(u\!\), "doubly outlined" \(v\!\), and "centrally dark" \(w\!\). We may regard these simple features as logical propositions \(u, v, w : X \to \mathbb{B}.\) The target concept \(\mathcal{Q}\) is one whose extension is a polymorphous set \(Q\!\), the subset \(Q\!\) of the universe \(X\!\) where the complex feature \(q : X \to \mathbb{B}\) holds true. The \(Q\!\) in question is defined by the requirement: "Having at least 2 of the 3 features in the set \(\{ u, v, w \}\!\)".
Taking the symbols \(u\!\) = "round", \(v\!\) = "doubly outlined", \(w\!\) = "centrally dark", and using the corresponding capital letters to label the circles of a venn diagram, we get a picture of the target set \(Q\!\) as the shaded region in Figure 1. Using these symbols as "sentence letters" in a truth table, let the truth function \(q\!\) mean the very same thing as the expression "(\(u\!\) and \(v\!\)) or (\(u\!\) and \(w\!\)) or (\(v\!\) and \(w\!\))".
o-----------------------------------------------------------o | X | | | | o-------------o | | / \ | | / \ | | / \ | | / \ | | / \ | | o o | | | U | | | | | | | | | | | | | | | | | | | o--o----------o o----------o--o | | / \%%%%%%%%%%\ /%%%%%%%%%%/ \ | | / \%%%%%%%%%%o%%%%%%%%%%/ \ | | / \%%%%%%%%/%\%%%%%%%%/ \ | | / \%%%%%%/%%%\%%%%%%/ \ | | / \%%%%/%%%%%\%%%%/ \ | | o o--o-------o--o o | | | |%%%%%%%| | | | | |%%%%%%%| | | | | |%%%%%%%| | | | | V |%%%%%%%| W | | | | |%%%%%%%| | | | o o%%%%%%%o o | | \ \%%%%%/ / | | \ \%%%/ / | | \ \%/ / | | \ o / | | \ / \ / | | o-------------o o-------------o | | | | | o-----------------------------------------------------------o Figure 1. Polymorphous Set Q
In other words, the proposition \(q\!\) is a truth-function of the 3 logical variables \(u\!\), \(v\!\), \(w\!\), and it may be evaluated according to the "truth table" scheme that is shown in Table 2. In this representation the polymorphous set \(Q\!\) appears in the guise of what some people call the "pre-image" or the "fiber of truth" under the function \(q\!\). More precisely, the 3-tuples for which \(q\!\) evaluates to true are in an obvious correspondence with the shaded cells of the venn diagram. No matter how we get down to the level of actual information, it's all pretty much the same stuff.
Table 2. Polymorphous Function q o---------------o-----------o-----------o-----------o-------o | u v w | u & v | u & w | v & w | q | o---------------o-----------o-----------o-----------o-------o | | | | | | | 0 0 0 | 0 | 0 | 0 | 0 | | | | | | | | 0 0 1 | 0 | 0 | 0 | 0 | | | | | | | | 0 1 0 | 0 | 0 | 0 | 0 | | | | | | | | 0 1 1 | 0 | 0 | 1 | 1 | | | | | | | | 1 0 0 | 0 | 0 | 0 | 0 | | | | | | | | 1 0 1 | 0 | 1 | 0 | 1 | | | | | | | | 1 1 0 | 1 | 0 | 0 | 1 | | | | | | | | 1 1 1 | 1 | 1 | 1 | 1 | | | | | | | o---------------o-----------o-----------o-----------o-------o
With the pictures of the venn diagram and the truth table before us, we have come to the verge of seeing how the word "model" is used in logic, namely, to distinguish whatever things satisfy a description.
In the venn diagram presentation, to be a model of some conceptual description \(\mathcal{F}\) is to be a point \(x\!\) in the corresponding region \(F\!\) of the universe of discourse \(X\!\).
In the truth table representation, to be a model of a logical proposition \(f\!\) is to be a data-vector \(\mathbf{x}\!\) (a row of the table) on which a function \(f\!\) evaluates to true.
This manner of speaking makes sense to those who consider the ultimate meaning of a sentence to be not the logical proposition that it denotes but its truth value instead. From the point of view, one says that any data-vector of this type (\(k\!\)-tuples of truth values) may be regarded as an "interpretation" of the proposition with \(k\!\) variables. An interpretation that yields a value of true is then called a "model".
For the most threadbare kind of logical system that we find residing in propositional calculus, this notion of model is almost too simple to deserve the name, yet it can be of service to fashion some form of continuity between the simple and the complex.
Note 3
The present is big with the future.
— Leibniz
Here I now delve into subject matters that are more specifically logical in the character of their interpretation.
Working Note. Need segue here to explain the use of Cactus Language.
Imagine that we are sitting in one of the cells of a venn diagram, contemplating the walls. There are \(k\!\) of them, one for each positive feature \(x_1, \ldots, x_k\) in our universe of discourse. Our particular cell is described by a concatenation of \(k\!\) signed assertions, positive or negative, regarding each of these features, and this description of our position amounts to what is called an "interpretation" of whatever proposition may rule the space, or reign on the universe of discourse. But are we locked into this interpretation?
With respect to each edge \(x\!\) of the cell we consider a test proposition \(\operatorname{d}x\) that determines our decision whether or not we will make a difference in how we stand regarding \(x\!\). If \(\operatorname{d}x\) is true then it marks our decision, intention, or plan to cross over the edge \(x\!\) at some point within the purview of the contemplated plan.
To reckon the effect of several such decisions on our current interpretation, or the value of the reigning proposition, we transform that position or that proposition by making the following array of substitutions everywhere in its expression:
\(1.\!\) Substitute "\((x_1, \operatorname{d}x_1)\)" for "\(x_1\!\)"
\(2.\!\) Substitute "\((x_2, \operatorname{d}x_2)\)" for "\(x_2\!\)"
\(3.\!\) Substitute "\((x_3, \operatorname{d}x_3)\)" for "\(x_3\!\)"
\(\ldots\)
\(k.\!\) Substitute "\((x_k, \operatorname{d}x_k)\)" for "\(x_k\!\)"
For concreteness, consider the polymorphous set \(Q\!\) of Example 1 and focus on the central cell, specifically, the cell described by the conjunction of logical features in the expression "\(u\ v\ w\)".
o-----------------------------------------------------------o | X | | | | o-------------o | | / \ | | / \ | | / \ | | / \ | | / \ | | o o | | | U | | | | | | | | | | | | | | | | | | | o--o----------o o----------o--o | | / \%%%%%%%%%%\ /%%%%%%%%%%/ \ | | / \%%%%%%%%%%o%%%%%%%%%%/ \ | | / \%%%%%%%%/%\%%%%%%%%/ \ | | / \%%%%%%/%%%\%%%%%%/ \ | | / \%%%%/%%%%%\%%%%/ \ | | o o--o-------o--o o | | | |%%%%%%%| | | | | |%%%%%%%| | | | | |%%%%%%%| | | | | V |%%%%%%%| W | | | | |%%%%%%%| | | | o o%%%%%%%o o | | \ \%%%%%/ / | | \ \%%%/ / | | \ \%/ / | | \ o / | | \ / \ / | | o-------------o o-------------o | | | | | o-----------------------------------------------------------o Figure 1. Polymorphous Set Q
The proposition or the truth-function \(q\!\) that describes \(Q\!\)is:
(( u v )( u w )( v w ))
Conjoining the query that specifies the center cell gives:
(( u v )( u w )( v w )) u v w
And we know the value of the interpretation by whether this last expression issues in a model.
Applying the enlargement operator \(\operatorname{E}\) to the initial proposition \(q\!\) yields:
(( ( u , du )( v , dv )
)( ( u , du )( w , dw )
)( ( v , dv )( w , dw )
))
Conjoining a query on the center cell yields:
(( ( u , du )( v , dv )
)( ( u , du )( w , dw )
)( ( v , dv )( w , dw )
))
u v w
The models of this last expression tell us which combinations of feature changes among the set \(\{ \operatorname{d}u, \operatorname{d}v, \operatorname{d}w \}\) will take us from our present interpretation, the center cell expressed by "\(u\ v\ w\)", to a true value under the target proposition (( u v )( u w )( v w )) .
The result of applying the difference operator \(\operatorname{D}\) to the initial proposition \(\operatorname{q}\), conjoined with a query on the center cell, yields:
(
(( ( u , du )( v , dv )
)( ( u , du )( w , dw )
)( ( v , dv )( w , dw )
))
,
(( u v
)( u w
)( v w
))
)
u v w
The models of this last proposition are:
1. u v w du dv dw
2. u v w du dv (dw)
3. u v w du (dv) dw
4. u v w (du) dv dw
This tells us that changing any two or more of the features \(u, v, w\) will take us from the center cell to a cell outside the shaded region for the set \(Q\!\).
Note 4
| It is one of the rules of my system of general harmony,
| 'that the present is big with the future', and that he
| who sees all sees in that which is that which shall be.
|
| Leibniz, 'Theodicy'
|
| Gottfried Wilhelm, Freiherr von Leibniz,
|'Theodicy: Essays on the Goodness of God,
| The Freedom of Man, & The Origin of Evil',
| Edited with an Introduction by Austin Farrer,
| Translated by E.M. Huggard from C.J. Gerhardt's
| Edition of the 'Collected Philosophical Works',
| 1875-90; Routledge & Kegan Paul, London, UK, 1951;
| Open Court, La Salle, IL, 1985. Paragraph 360, Page 341.
To round out the presentation of the "Polymorphous" Example 1,
I will go through what has gone before and lay in the graphic
forms of all of the propositional expressions. These graphs,
whose official botanical designation makes them out to be
a species of "painted and rooted cacti" (PARC's), are not
too far from the actual graph-theoretic data-structures
that result from parsing the Cactus string expressions,
the "painted and rooted cactus expressions" (PARCE's).
Finally, I will add a couple of venn diagrams that
will serve to illustrate the "difference opus" Dq.
If you apply an operator to an operand you must
arrive at either an opus or an opera, no?
Consider the polymorphous set Q of Example 1 and focus on the central cell,
described by the conjunction of logical features in the expression "u v w".
o-------------------------------------------------o
| X |
| |
| o-------------o |
| / \ |
| / \ |
| / \ |
| / \ |
| o U o |
| | | |
| | | |
| | | |
| o---o---------o o---------o---o |
| / \%%%%%%%%%\ /%%%%%%%%%/ \ |
| / \%%%%%%%%%o%%%%%%%%%/ \ |
| / \%%%%%%%/%\%%%%%%%/ \ |
| / \%%%%%/%%%\%%%%%/ \ |
| o o---o-----o---o o |
| | |%%%%%| | |
| | V |%%%%%| W | |
| | |%%%%%| | |
| o o%%%%%o o |
| \ \%%%/ / |
| \ \%/ / |
| \ o / |
| \ / \ / |
| o-------------o o-------------o |
| |
| |
o-------------------------------------------------o
Figure 1. Polymorphous Set Q
The proposition or truth-function q : X -> B that
describes Q is represented by the following graph
and text expressions:
o-------------------------------------------------o
| q |
o-------------------------------------------------o
| |
| u v u w v w |
| o o o |
| \ | / |
| \ | / |
| \|/ |
| o |
| | |
| | |
| | |
| @ |
| |
o-------------------------------------------------o
| (( u v )( u w )( v w )) |
o-------------------------------------------------o
Conjoining the query that specifies the center cell gives:
o-------------------------------------------------o
| q.uvw |
o-------------------------------------------------o
| |
| u v u w v w |
| o o o |
| \ | / |
| \ | / |
| \|/ |
| o |
| | |
| | |
| | |
| @ u v w |
| |
o-------------------------------------------------o
| (( u v )( u w )( v w )) u v w |
o-------------------------------------------------o
And we know the value of the interpretation by
whether this last expression issues in a model.
Applying the enlargement operator E
to the initial proposition q yields:
o-------------------------------------------------o
| Eq |
o-------------------------------------------------o
| |
| u du v dv u du w dw v dv w dw |
| o---o o---o o---o o---o o---o o---o |
| \ | | / \ | | / \ | | / |
| \ | | / \ | | / \ | | / |
| \| |/ \| |/ \| |/ |
| o=o o=o o=o |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \|/ |
| o |
| | |
| | |
| | |
| @ |
| |
o-------------------------------------------------o
| |
| (( ( u , du ) ( v , dv ) |
| )( ( u , du ) ( w , dw ) |
| )( ( v , dv ) ( w , dw ) |
| )) |
| |
o-------------------------------------------------o
Conjoining a query on the center cell yields:
o-------------------------------------------------o
| Eq.uvw |
o-------------------------------------------------o
| |
| u du v dv u du w dw v dv w dw |
| o---o o---o o---o o---o o---o o---o |
| \ | | / \ | | / \ | | / |
| \ | | / \ | | / \ | | / |
| \| |/ \| |/ \| |/ |
| o=o o=o o=o |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \|/ |
| o |
| | |
| | |
| | |
| @ u v w |
| |
o-------------------------------------------------o
| |
| (( ( u , du ) ( v , dv ) |
| )( ( u , du ) ( w , dw ) |
| )( ( v , dv ) ( w , dw ) |
| )) |
| |
| u v w |
| |
o-------------------------------------------------o
The models of this last expression tell us which combinations of
feature changes among the set {du, dv, dw} will take us from our
present interpretation, the center cell expressed by "u v w", to
a true value under the target proposition (( u v )( u w )( v w )).
The result of applying the difference operator D
to the initial proposition q, conjoined with
a query on the center cell, yields:
o-------------------------------------------------o
| Dq.uvw |
o-------------------------------------------------o
| |
| u du v dv u du w dw v dv w dw |
| o---o o---o o---o o---o o---o o---o |
| \ | | / \ | | / \ | | / |
| \ | | / \ | | / \ | | / |
| \| |/ \| |/ \| |/ |
| o=o o=o o=o |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \ | / u v u w v w |
| \ | / o o o |
| \ | / \ | / |
| \ | / \ | / |
| \|/ \|/ |
| o o |
| | | |
| | | |
| | | |
| o---------------o |
| \ / |
| \ / |
| \ / |
| \ / |
| \ / |
| \ / |
| \ / |
| @ u v w |
| |
o-------------------------------------------------o
| |
| ( |
| (( ( u , du ) ( v , dv ) |
| )( ( u , du ) ( w , dw ) |
| )( ( v , dv ) ( w , dw ) |
| )) |
| , |
| (( u v |
| )( u w |
| )( v w |
| )) |
| ) |
| |
| u v w |
| |
o-------------------------------------------------o
The models of this last proposition are:
1. u v w du dv dw
2. u v w du dv (dw)
3. u v w du (dv) dw
4. u v w (du) dv dw
This tells us that changing any two or more of the
features u, v, w will take us from the center cell,
as described by the conjunctive expression "u v w",
to a cell outside the shaded region for the set Q.
o-------------------------------------------------o
| X |
| |
| o-------------o |
| / \ |
| / U \ |
| / \ |
| / \ |
| o @ o |
| | ^ | |
| | |dw | |
| | | | @ |
| o---o---------o o----|----o---o ^ |
| / \`````````\ /`````|```/ \ /dw |
| / du \`````dw``o``dv``|``/ \/ |
| / @<-----\-o<----/+\---->o`/ /\ |
| / \`````/`|`\`````/ / \ |
| o o---o--|--o---o / o |
| | |``|``| / | |
| | V |`du``| / W | |
| | |` |``| / | |
| o o``v``o dv / o |
| \ \`o-/------->@ / |
| \ \`/ / |
| \ o / |
| \ / \ / |
| o-------------o o-------------o |
| |
| |
o-------------------------------------------------o
Figure 3. Effect of the Difference Operator D
Acting on a Polymorphous Function q
Figure 3 shows one way to picture this kind of a situation,
by superimposing the paths of indicated feature changes on
the venn diagram of the underlying proposition. Here, the
models, or the satisfying interpretations, of the relevant
"difference proposition" Dq are marked with "@" signs, and
the boundary crossings along each path are marked with the
corresponding "differential features" among the collection
{du, dv, dw}. In sum, starting from the cell uvw, we have
the following four paths:
1. du dv dw => Change u, v, w.
2. du dv (dw) => Change u and v.
3. du (dv) dw => Change u and w.
4. (du) dv dw => Change v and w.
Next I will discuss several applications of logical differentials,
developing along the way their logical and practical implications.
Note 5
We have come to the point of making a connection,
at a very primitive level, between propositional
logic and the classes of mathematical structures
that are employed in mathematical systems theory
to model dynamical systems of very general sorts.
Here is a flash montage of what has gone before,
retrospectively touching on just the highpoints,
and highlighting mostly just Figures and Tables,
all directed toward the aim of ending up with a
novel style of pictorial diagram, one that will
serve us well in the future, as I have found it
readily adaptable and steadily more trustworthy
in my previous investigations, whenever we have
to illustrate these very basic sorts of dynamic
scenarios to ourselves, to others, to computers.
We typically start out with a proposition of interest,
for example, the proposition q : X -> B depicted here:
o-------------------------------------------------o
| q |
o-------------------------------------------------o
| |
| u v u w v w |
| o o o |
| \ | / |
| \ | / |
| \|/ |
| o |
| | |
| | |
| | |
| @ |
| |
o-------------------------------------------------o
| (( u v )( u w )( v w )) |
o-------------------------------------------------o
The proposition q is properly considered as an "abstract object",
in some acceptation of those very bedevilled and egging-on terms,
but it enjoys an interpretation as a function of a suitable type,
and all we have to do in order to enjoy the utility of this type
of representation is to observe a decent respect for what befits.
I will skip over the details of how to do this for right now.
I started to write them out in full, and it all became even
more tedious than my usual standard, and besides, I think
that everyone more or less knows how to do this already.
Once we have survived the big leap of re-interpreting these
abstract names as the names of relatively concrete dimensions
of variation, we can begin to lay out all of the familiar sorts
of mathematical models and pictorial diagrams that go with these
modest dimensions, the functions that can be formed on them, and
the transformations that can be entertained among this whole crew.
Here is the venn diagram for the proposition q.
o-----------------------------------------------------------o
| X |
| |
| o-------------o |
| / \ |
| / \ |
| / \ |
| / \ |
| / \ |
| o o |
| | U | |
| | | |
| | | |
| | | |
| | | |
| o--o----------o o----------o--o |
| / \%%%%%%%%%%\ /%%%%%%%%%%/ \ |
| / \%%%%%%%%%%o%%%%%%%%%%/ \ |
| / \%%%%%%%%/%\%%%%%%%%/ \ |
| / \%%%%%%/%%%\%%%%%%/ \ |
| / \%%%%/%%%%%\%%%%/ \ |
| o o--o-------o--o o |
| | |%%%%%%%| | |
| | |%%%%%%%| | |
| | |%%%%%%%| | |
| | V |%%%%%%%| W | |
| | |%%%%%%%| | |
| o o%%%%%%%o o |
| \ \%%%%%/ / |
| \ \%%%/ / |
| \ \%/ / |
| \ o / |
| \ / \ / |
| o-------------o o-------------o |
| |
| |
o-----------------------------------------------------------o
Figure 1. Venn Diagram for the Proposition q
By way of excuse, if not yet a full justification, I probably ought to give
an account of the reasons why I continue to hang onto these primitive styles
of depiction, even though I can hardly recommend that anybody actually try to
draw them, at least, not once the number of variables climbs much higher than
three or four or five at the utmost. One of the reasons would have to be this:
that in the relationship between their continuous aspect and their discrete aspect,
venn diagrams constitute a form of "iconic" reminder of a very important fact about
all "finite information depictions" (FID's) of the larger world of reality, and that
is the hard fact that we deceive ourselves to a degree if we imagine that the lines
and the distinctions that we draw in our imagination are all there is to reality,
and thus, that as we practice to categorize, we also manage to discretize, and
thus, to distort, to reduce, and to truncate the richness of what there is to
the poverty of what we can sieve and sift through our senses, or what we can
draw in the tangled webs of our own very tenuous and tinctured distinctions.
Another common scheme for description and evaluation of a proposition
is the so-called "truth table" or the "semantic tableau", for example:
Table 2. Truth Table for the Proposition q
o---------------o-----------o-----------o-----------o-------o
| u v w | u & v | u & w | v & w | q |
o---------------o-----------o-----------o-----------o-------o
| | | | | |
| 0 0 0 | 0 | 0 | 0 | 0 |
| | | | | |
| 0 0 1 | 0 | 0 | 0 | 0 |
| | | | | |
| 0 1 0 | 0 | 0 | 0 | 0 |
| | | | | |
| 0 1 1 | 0 | 0 | 1 | 1 |
| | | | | |
| 1 0 0 | 0 | 0 | 0 | 0 |
| | | | | |
| 1 0 1 | 0 | 1 | 0 | 1 |
| | | | | |
| 1 1 0 | 1 | 0 | 0 | 1 |
| | | | | |
| 1 1 1 | 1 | 1 | 1 | 1 |
| | | | | |
o---------------o-----------o-----------o-----------o-------o
Reading off the shaded cells of the venn diagram or the
rows of the truth table that have a "1" in the q column,
we see that the "models", or satisfying interpretations,
of the proposition q are the four that can be expressed,
in either the "additive" or the "multiplicative" manner,
as follows:
1. The points of the space X that are assigned the coordinates:
<u, v, w> = <0, 1, 1> or <1, 0, 1> or <1, 1, 0> or <1, 1, 1>.
2. The points of the space X that have the conjunctive descriptions:
"(u) v w", "u (v) w", "u v (w)", "u v w", where "(x)" is "not x".
The next thing that one typically does is to consider the effects
of various "operators" on the proposition of interest, which may
be called the "operand" or the "source" proposition, leaving the
corresponding terms "opus" or "target" as names for the result.
In our initial consideration of the proposition q, we naturally
interpret it as a function of the three variables that it wears
on its sleeve, as it were, namely, those that we find contained
in the basis {u, v, w}. As we begin to regard this proposition
from the standpoint of a differential analysis, however, we may
need to regard it as "tacitly embedded" in any number of higher
dimensional spaces. Just by way of starting out, our immediate
interest is with the "first order differential analysis" (FODA),
and this requires us to regard all of the propositions in sight
as functions of the variables in the first order extended basis,
specifically, those in the set {u, v, w, du, dv, dw}. Now this
does not change the expression of any proposition, like q, that
does not mention the extra variables, only changing how it gets
interpreted as a function. A level of interpretive flexibility
of this order is very useful, and it is quite common throughout
mathematics. In this discussion, I will invoke its application
under the name of the "tacit extension" of a proposition to any
universe of discourse based on a superset of its original basis.
Note 6
I think that we finally have enough of the preliminary
set-ups and warm-ups out of the way that we can begin
to tackle the differential analysis proper of our
sample proposition q = (( u v )( u w )( v w )).
When X is the type of space that is generated by {u, v, w},
let dX be the type of space that is generated by (du, dv, dw},
and let X x dX be the type of space that is generated by the
extended set of boolean basis elements {u, v, w, du, dv, dw}.
For convenience, define a notation "EX" so that EX = X x dX.
Even though the differential variables are in some abstract
sense no different than other boolean variables, it usually
helps to mark their distinctive roles and their differential
interpretation by means of the distinguishing domain name "dB".
Using these designations of logical spaces, the propositions
over them can be assigned both abstract and concrete types.
For instance, consider the proposition q<u, v, w>, as before,
and then consider its tacit extension q<u, v, w, du, dv, dw>,
the latter of which may be indicated more explicitly as "eq".
1. Proposition q is abstractly typed as q : B^3 -> B.
Proposition q is concretely typed as q : X -> B.
2. Proposition eq is abstractly typed as eq : B^3 x dB^3 -> B.
Proposition eq is concretely typed as eq : X x dX -> B.
Succinctly, eq : EX -> B.
We now return to our consideration of the effects
of various differential operators on propositions.
This time around we have enough exact terminology
that we shall be able to explain what is actually
going on here in a rather more articulate fashion.
The first transformation of the source proposition q that we may
wish to stop and examine, though it is not unusual to skip right
over this stage of analysis, frequently regarding it as a purely
intermediary stage, holding scarcely even so much as the passing
interest, is the work of the "enlargement" or "shift" operator E.
Applying the operator E to the operand proposition q yields:
o-------------------------------------------------o
| Eq |
o-------------------------------------------------o
| |
| u du v dv u du w dw v dv w dw |
| o---o o---o o---o o---o o---o o---o |
| \ | | / \ | | / \ | | / |
| \ | | / \ | | / \ | | / |
| \| |/ \| |/ \| |/ |
| o=o o=o o=o |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \|/ |
| o |
| | |
| | |
| | |
| @ |
| |
o-------------------------------------------------o
| |
| (( ( u , du ) ( v , dv ) |
| )( ( u , du ) ( w , dw ) |
| )( ( v , dv ) ( w , dw ) |
| )) |
| |
o-------------------------------------------------o
The enlarged proposition Eq is a minimally interpretable as
as a function on the six variables of {u, v, w, du, dv, dw}.
In other words, Eq : EX -> B, or Eq : X x dX -> B.
Conjoining a query on the center cell, c = uvw, yields:
o-------------------------------------------------o
| Eq.c |
o-------------------------------------------------o
| |
| u du v dv u du w dw v dv w dw |
| o---o o---o o---o o---o o---o o---o |
| \ | | / \ | | / \ | | / |
| \ | | / \ | | / \ | | / |
| \| |/ \| |/ \| |/ |
| o=o o=o o=o |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \|/ |
| o |
| | |
| | |
| | |
| @ u v w |
| |
o-------------------------------------------------o
| |
| (( ( u , du ) ( v , dv ) |
| )( ( u , du ) ( w , dw ) |
| )( ( v , dv ) ( w , dw ) |
| )) |
| |
| u v w |
| |
o-------------------------------------------------o
The models of this last expression tell us which combinations of
feature changes among the set {du, dv, dw} will take us from our
present interpretation, the center cell expressed by "u v w", to
a true value under the given proposition (( u v )( u w )( v w )).
The models of Eq.c can be described in the usual ways as follows:
1. The points of the space EX that have
the following coordinate descriptions:
<u, v, w, du, dv, dw> =
<1, 1, 1, 0, 0, 0>,
<1, 1, 1, 0, 0, 1>,
<1, 1, 1, 0, 1, 0>,
<1, 1, 1, 1, 0, 0>.
2. The points of the space EX that have
the following conjunctive expressions:
u v w (du)(dv)(dw),
u v w (du)(dv) dw ,
u v w (du) dv (dw),
u v w du (dv)(dw).
In summary, Eq.c informs us that we can get from c to a model of q by
making the following changes in our position with respect to u, v, w,
to wit, "change none or just one among {u, v, w}".
I think that it would be worth our time to diagram the models
of the "enlarged" or "shifted" proposition, Eq, at least, the
selection of them that we find issuing from the center cell c.
Figure 4 is an extended venn diagram for the proposition Eq.c,
where the shaded area gives the models of q and the "@" signs
mark the terminal points of the requisite feature alterations.
o-------------------------------------------------o
| X |
| |
| o-------------o |
| / \ |
| / \ |
| / \ |
| / \ |
| o U o |
| | | |
| | | |
| | | |
| o---o---------o o---------o---o |
| / \`````````\ /`````````/ \ |
| / \`````dw``o``dv`````/ \ |
| / \`@<----/@\---->@`/ \ |
| / \`````/`|`\`````/ \ |
| o o---o--|--o---o o |
| | |``|``| | |
| | V |`du``| W | |
| | |` |``| | |
| o o``v``o o |
| \ \`@`/ / |
| \ \`/ / |
| \ o / |
| \ / \ / |
| o-------------o o-------------o |
| |
| |
o-------------------------------------------------o
Figure 4. Effect of the Enlargement Operator E
On the Proposition q, Evaluated at c
Note 7
One more piece of notation will save us a few bytes
in the length of many of our schematic formulations.
Let !X! = {x_1, ..., x_k} be a finite class of variables --
whose names I list, according to the usual custom, without
what seems to my semiotic consciousness like the necessary
quotation marks around their particular characters, though
not without not a little trepidation, or without a worried
cognizance that I may be obligated to reinsert them all to
their rightful places at a subsequent stage of development --
with regard to which we may now define the following items:
1. The "(first order) differential alphabet",
d!X! = {dx_1, ..., dx_k}.
2. The "(first order) extended alphabet",
E!X! = !X! |_| d!X!,
E!X! = {x_1, ..., x_k, dx_1, ..., dx_k}.
Before we continue with the differential analysis
of the source proposition q, we need to pause and
take another look at just how it shapes up in the
light of the extended universe EX, in other words,
to examine in utter detail its tacit extension eq.
The models of eq in EX can be comprehended as follows:
1. Working in the "summary coefficient" form of representation,
if the coordinate list x is a model of q in X, then one can
construct a coordinate list ex as a model for eq in EX just
by appending any combination of values for the differential
variables in d!X!.
For example, to focus once again on the center cell c,
which happens to be a model of the proposition q in X,
one can extend c in eight different ways into EX, and
thus get eight models of the tacit extension eq in EX.
Though it may seem an utter triviality to write these
out, I will do it for the sake of seeing the patterns.
The models of eq in EX that are tacit extensions of c:
<u, v, w, du, dv, dw> =
<1, 1, 1, 0, 0, 0>,
<1, 1, 1, 0, 0, 1>,
<1, 1, 1, 0, 1, 0>,
<1, 1, 1, 0, 1, 1>,
<1, 1, 1, 1, 0, 0>,
<1, 1, 1, 1, 0, 1>,
<1, 1, 1, 1, 1, 0>,
<1, 1, 1, 1, 1, 1>.
2. Working in the "conjunctive product" form of representation,
if the conjunct symbol x is a model of q in X, then one can
construct a conjunct symbol ex as a model for eq in EX just
by appending any combination of values for the differential
variables in d!X!.
The models of eq in EX that are tacit extensions of c:
u v w (du)(dv)(dw),
u v w (du)(dv) dw ,
u v w (du) dv (dw),
u v w (du) dv dw ,
u v w du (dv)(dw),
u v w du (dv) dw ,
u v w du dv (dw),
u v w du dv dw .
In short, eq.c just enumerates all of the possible changes in EX
that "derive from", "issue from", or "stem from" the cell c in X.
Okay, that was pretty tedious, and I know that it all appears
to be totally trivial, which is precisely why we usually just
leave it "tacit" in the first place, but hard experience, and
a real acquaintance with the confusion that can beset us when
we do not render these implicit grounds explicit, have taught
me that it will ultimately be necessary to get clear about it,
and by this "clear" to say "marked", not merely "transparent".
Note 8
Before going on -- in order to keep alive the will to go on! --
it would probably be a good idea to remind ourselves of just
why we are going through with this exercise. It is to unify
the world of change, for which aspect or regime of the world
I occasionally evoke the eponymous figures of Prometheus and
Heraclitus, and the world of logic, for which facet or realm
of the world I periodically recur to the prototypical shades
of Epimetheus and Parmenides, at least, that is, to state it
more carefully, to encompass the antics and the escapades of
these all too manifestly strife-born twins within the scopes
of our thoughts and within the charts of our theories, as it
is most likely the only places where ever they will, for the
moment and as long as it lasts, be seen or be heard together.
With that intermezzo, with all of its echoes of the opening overture,
over and done, let us now return to that droller drama, already fast
in progress, the differential disentanglements, hopefully toward the
end of a grandly enlightening denouement, of the ever-polymorphous Q.
The next transformation of the source proposition q, that we are
typically aiming to contemplate in the process of carrying out a
"differential analysis" of its "dynamic" effects or implications,
is the yield of the so-called "difference" or "delta" operator D.
The resultant "difference proposition" Dq is defined in terms of
the source proposition q and the "shifted proposition" Eq thusly:
| Dq = Eq - q = Eq - eq.
|
| Since "+" and "-" signify the same operation over B, we have:
|
| Dq = Eq + q = Eq + eq.
|
| Since "+" = "exclusive-or", RefLog syntax expresses this as:
|
| Eq q Eq eq
| o---o o---o
| \ / \ /
| Dq = @ = @
|
| Dq = ( Eq , q ) = ( Eq , eq ).
|
| Recall that a k-place bracket "(x_1, x_2, ..., x_k)"
| is interpreted (in the "existential interpretation")
| to mean "Exactly one of the x_j is false", thus the
| two-place bracket is equivalent to the exclusive-or.
The result of applying the difference operator D to the source
proposition q, conjoined with a query on the center cell c, is:
o-------------------------------------------------o
| Dq.uvw |
o-------------------------------------------------o
| |
| u du v dv u du w dw v dv w dw |
| o---o o---o o---o o---o o---o o---o |
| \ | | / \ | | / \ | | / |
| \ | | / \ | | / \ | | / |
| \| |/ \| |/ \| |/ |
| o=o o=o o=o |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \ | / |
| \ | / u v u w v w |
| \ | / o o o |
| \ | / \ | / |
| \ | / \ | / |
| \|/ \|/ |
| o o |
| | | |
| | | |
| | | |
| o---------------o |
| \ / |
| \ / |
| \ / |
| \ / |
| \ / |
| \ / |
| \ / |
| @ u v w |
| |
o-------------------------------------------------o
| |
| ( |
| (( ( u , du ) ( v , dv ) |
| )( ( u , du ) ( w , dw ) |
| )( ( v , dv ) ( w , dw ) |
| )) |
| , |
| (( u v |
| )( u w |
| )( v w |
| )) |
| ) |
| |
| u v w |
| |
o-------------------------------------------------o
The models of the difference proposition Dq.uvw are:
1. u v w du dv dw
2. u v w du dv (dw)
3. u v w du (dv) dw
4. u v w (du) dv dw
This tells us that changing any two or more of the
features u, v, w will take us from the center cell
that is marked by the conjunctive expression "uvw",
to a cell outside the shaded region for the area Q.
o-------------------------------------------------o
| X |
| |
| o-------------o |
| / \ |
| / U \ |
| / \ |
| / \ |
| o @ o |
| | ^ | |
| | |dw | |
| | | | @ |
| o---o---------o o----|----o---o ^ |
| / \`````````\ /`````|```/ \ /dw |
| / du \`````dw``o``dv``|``/ \/ |
| / @<-----\-o<----/+\---->o`/ /\ |
| / \`````/`|`\`````/ / \ |
| o o---o--|--o---o / o |
| | |``|``| / | |
| | V |`du``| / W | |
| | |` |``| / | |
| o o``v``o dv / o |
| \ \`o-/------->@ / |
| \ \`/ / |
| \ o / |
| \ / \ / |
| o-------------o o-------------o |
| |
| |
o-------------------------------------------------o
Figure 3. Effect of the Difference Operator D
Acting on a Polymorphous Function q
Figure 3 shows one way to picture this kind of a situation,
by superimposing the paths of indicated feature changes on
the venn diagram of the underlying proposition. Here, the
models, or the satisfying interpretations, of the relevant
"difference proposition" Dq are marked with "@" signs, and
the boundary crossings along each path are marked with the
corresponding "differential features" among the collection
{du, dv, dw}. In sum, starting from the cell uvw, we have
the following four paths:
1. du dv dw = Change u, v, w.
2. du dv (dw) = Change u and v.
3. du (dv) dw = Change u and w.
4. (du) dv dw = Change v and w.
That sums up, but rather more carefully, the material that
I ran through just a bit too quickly the first time around.
Next time, I will begin to develop an alternative style of
diagram for depicting these types of differential settings.
Note 9
Another way of looking at this situation is by letting the (first order) differential features du, dv, dw be viewed as the features of another universe of discourse, called the "tangent universe to X with respect to the interpretation c" and represented as dX.c. In this setting, Dq.c, the "difference proposition of q at the interpretation c", where c = uvw, is marked by the shaded region in Figure 4. o-----------------------------------------------------------o | dX.c | | | | o-------------o | | / \ | | / \ | | / \ | | / \ | | / \ | | o o | | | dU | | | | | | | | | | | | | | | | | | | o--o----------o o----------o--o | | / \``````````\ /``````````/ \ | | / \````2`````o`````3````/ \ | | / \````````/`\````````/ \ | | / \``````/```\``````/ \ | | / \````/``1``\````/ \ | | o o--o-------o--o o | | | |```````| | | | | |```````| | | | | |```````| | | | | dV |```4```| dW | | | | |```````| | | | o o```````o o | | \ \`````/ / | | \ \```/ / | | \ \`/ / | | \ o / | | \ / \ / | | o-------------o o-------------o | | | | | o-----------------------------------------------------------o Figure 4. Tangent Venn Diagram for Dq.c Taken in the context of the tangent universe to X at c = uvw, written dX.c or dX.uvw, the shaded area of Figure 4 indicates the models of the difference proposition Dq.uvw, specifically: 1. u v w du dv dw 2. u v w du dv (dw) 3. u v w du (dv) dw 4. u v w (du) dv dw
Note 10
Sub*Title. There's Gonna Be A Rumble Tonight! From: "Theme One: A Program of Inquiry", Jon Awbrey & Susan Awbrey, August 9, 1989. Example 5. Jets and Sharks The propositional calculus that is based on the boundary operator can be interpreted in a way that resembles the logic of activation states and competition constraints in certain neural network models. One way to do this is by interpreting the blank or unmarked state as the resting state of a neural pool, the bound or marked state as its activated state, and by representing a mutually inhibitory pool of neurons A, B, C in the expression "(A, B, C)". To illustrate this possibility, we transcribe a well-known example from the parallel distributed processing literature (McClelland & Rumelhart, 1988) and work through two of the associated exercises as portrayed in Existential Graph format. File "jas.log". Jets and Sharks Example o-----------------------------------------------------------o | | | (( art ),( al ),( sam ),( clyde ),( mike ), | | ( jim ),( greg ),( john ),( doug ),( lance ), | | ( george ),( pete ),( fred ),( gene ),( ralph ), | | ( phil ),( ike ),( nick ),( don ),( ned ), | | ( karl ),( ken ),( earl ),( rick ),( ol ), | | ( neal ),( dave )) | | | | ( jets , sharks ) | | | | ( jets , | | ( art ),( al ),( sam ),( clyde ),( mike ), | | ( jim ),( greg ),( john ),( doug ),( lance ), | | ( george ),( pete ),( fred ),( gene ),( ralph )) | | | | ( sharks , | | ( phil ),( ike ),( nick ),( don ),( ned ),( karl ), | | ( ken ),( earl ),( rick ),( ol ),( neal ),( dave )) | | | | (( 20's ),( 30's ),( 40's )) | | | | ( 20's , | | ( sam ),( jim ),( greg ),( john ),( lance ), | | ( george ),( pete ),( fred ),( gene ),( ken )) | | | | ( 30's , | | ( al ),( mike ),( doug ),( ralph ),( phil ), | | ( ike ),( nick ),( don ),( ned ),( rick ), | | ( ol ),( neal ),( dave )) | | | | ( 40's , | | ( art ),( clyde ),( karl ),( earl )) | | | | (( junior_high ),( high_school ),( college )) | | | | ( junior_high , | | ( art ),( al ),( clyde ),( mike ),( jim ), | | ( john ),( lance ),( george ),( ralph ),( ike )) | | | | ( high_school , | | ( greg ),( doug ),( pete ),( fred ), | | ( nick ),( karl ),( ken ),( earl ), | | ( rick ),( neal ),( dave )) | | | | ( college , | | ( sam ),( gene ),( phil ),( don ),( ned ),( ol )) | | | | (( single ),( married ),( divorced )) | | | | ( single , | | ( art ),( sam ),( clyde ),( mike ),( doug ), | | ( pete ),( fred ),( gene ),( ralph ),( ike ), | | ( nick ),( ken ),( neal )) | | | | ( married , | | ( al ),( greg ),( john ),( lance ),( phil ), | | ( don ),( ned ),( karl ),( earl ),( ol )) | | | | ( divorced , | | ( jim ),( george ),( rick ),( dave )) | | | | (( bookie ),( burglar ),( pusher )) | | | | ( bookie , | | ( sam ),( clyde ),( mike ),( doug ), | | ( pete ),( ike ),( ned ),( karl ),( neal )) | | | | ( burglar , | | ( al ),( jim ),( john ),( lance ), | | ( george ),( don ),( ken ),( earl ),( rick )) | | | | ( pusher , | | ( art ),( greg ),( fred ),( gene ), | | ( ralph ),( phil ),( nick ),( ol ),( dave )) | | | o-----------------------------------------------------------o We now apply 'Study' to the proposition defining the Jets and Sharks data base. With a query on the name "ken" we obtain the following output, giving all the features associated with Ken: File "ken.sen". Output of Query on "ken" o-----------------------------------------------------------o | | | ken | | sharks | | 20's | | high_school | | single | | burglar | | | o-----------------------------------------------------------o With a query on the two features "college" and "sharks" we obtain the following outline of all features satisfying these constraints: File "cos.sen". Output of Query on "college" and "sharks" o-----------------------------------------------------------o | | | college | | sharks | | 30's | | married | | bookie | | ned | | burglar | | don | | pusher | | phil | | ol | | | o-----------------------------------------------------------o From this we discover that all college Sharks are 30-something and married. Further, we have a complete listing of their names broken down by occupation, as no doubt all of them will be, eventually. Reference. | McClelland, James L. & Rumelhart, David E., |'Explorations in Parallel Distributed Processing: | A Handbook of Models, Programs, and Exercises', | MIT Press, Cambridge, MA, 1988. Those who already know the tune, Be at liberty to sing out of it.
Note 11
| "The burden of genius is undeliverable" | From a poster, as I once misread it, | Marlboro, Vermont, c. 1976 How does Cosmo, and by this I mean my pet personification of cosmic order in the universe, not to be too tautologous about it, preserve a memory like that, a goodly fraction of a century later, whether localized to this body that's kept going by this heart, and whether by common assumption still more localized to the spongey fibres of this brain, or not? It strikes me, as it has struck others, that it's terribly unlikely to be stored in persistent patterns of activation, for "activation" and "persistent" are nigh a contradiction in terms, as even the author, Cosmo, of the 'I Ching' knew. But that was then, this is now, so let me try to say it planar.
Note 12
I happened on the graphical syntax for propositional calculus that
I now call the "cactus language" while exploring the confluence of
three streams of thought. There was C.S. Peirce's use of operator
variables in logical forms and the operational representations of
logical concepts, there was George Spencer Brown's explanation of
a variable as the contemplated presence or absence of a constant,
and then there was the graph theory and group theory that I had
been picking up, bit by bit, since I first encountered them in
tandem in Frank Harary's foundations of math course, c. 1970.
More on that later, as the memories unthaw, but for the moment
I want very much to take care of some long-unfinished business,
and give a more detailed explanation of how I used this syntax
to represent a popular exercise from the PDP literature of the
late 1980's, McClelland's and Rumelhart's "Jets and Sharks".
The knowledge base of the case can be expressed as a single proposition.
The following display presents it in the corresponding text file format.
File "jas.log". Jets and Sharks Example
o-----------------------------------------------------------o
| |
| (( art ),( al ),( sam ),( clyde ),( mike ), |
| ( jim ),( greg ),( john ),( doug ),( lance ), |
| ( george ),( pete ),( fred ),( gene ),( ralph ), |
| ( phil ),( ike ),( nick ),( don ),( ned ), |
| ( karl ),( ken ),( earl ),( rick ),( ol ), |
| ( neal ),( dave )) |
| |
| ( jets , sharks ) |
| |
| ( jets , |
| ( art ),( al ),( sam ),( clyde ),( mike ), |
| ( jim ),( greg ),( john ),( doug ),( lance ), |
| ( george ),( pete ),( fred ),( gene ),( ralph )) |
| |
| ( sharks , |
| ( phil ),( ike ),( nick ),( don ),( ned ),( karl ), |
| ( ken ),( earl ),( rick ),( ol ),( neal ),( dave )) |
| |
| (( 20's ),( 30's ),( 40's )) |
| |
| ( 20's , |
| ( sam ),( jim ),( greg ),( john ),( lance ), |
| ( george ),( pete ),( fred ),( gene ),( ken )) |
| |
| ( 30's , |
| ( al ),( mike ),( doug ),( ralph ),( phil ), |
| ( ike ),( nick ),( don ),( ned ),( rick ), |
| ( ol ),( neal ),( dave )) |
| |
| ( 40's , |
| ( art ),( clyde ),( karl ),( earl )) |
| |
| (( junior_high ),( high_school ),( college )) |
| |
| ( junior_high , |
| ( art ),( al ),( clyde ),( mike ),( jim ), |
| ( john ),( lance ),( george ),( ralph ),( ike )) |
| |
| ( high_school , |
| ( greg ),( doug ),( pete ),( fred ), |
| ( nick ),( karl ),( ken ),( earl ), |
| ( rick ),( neal ),( dave )) |
| |
| ( college , |
| ( sam ),( gene ),( phil ),( don ),( ned ),( ol )) |
| |
| (( single ),( married ),( divorced )) |
| |
| ( single , |
| ( art ),( sam ),( clyde ),( mike ),( doug ), |
| ( pete ),( fred ),( gene ),( ralph ),( ike ), |
| ( nick ),( ken ),( neal )) |
| |
| ( married , |
| ( al ),( greg ),( john ),( lance ),( phil ), |
| ( don ),( ned ),( karl ),( earl ),( ol )) |
| |
| ( divorced , |
| ( jim ),( george ),( rick ),( dave )) |
| |
| (( bookie ),( burglar ),( pusher )) |
| |
| ( bookie , |
| ( sam ),( clyde ),( mike ),( doug ), |
| ( pete ),( ike ),( ned ),( karl ),( neal )) |
| |
| ( burglar , |
| ( al ),( jim ),( john ),( lance ), |
| ( george ),( don ),( ken ),( earl ),( rick )) |
| |
| ( pusher , |
| ( art ),( greg ),( fred ),( gene ), |
| ( ralph ),( phil ),( nick ),( ol ),( dave )) |
| |
o-----------------------------------------------------------o
Let's start with the simplest clause of the conjoint proposition:
( jets , sharks )
Drawn as the corresponding cactus graph, we have:
jets sharks
o-----o
\ /
\ /
@
According to my earlier, if somewhat sketchy interpretive suggestions,
we are supposed to picture a quasi-neural pool that contains a couple
of quasi-neural agents or "units", that between the two of them stand
for the logical variables "jets" and "sharks", respectively. Further,
we imagine these agents to be mutually inhibitory, so that settlement
of the dynamic between them achieves equilibrium when just one of the
two is "active" or "changing" and the other is "stable" or "enduring".
Note 13
We were focussing on a particular figure of syntax,
presented here in both graph and string renditions:
o-------------------------------------------------o
| |
| x y |
| o-----o |
| \ / |
| \ / |
| @ |
| |
o-------------------------------------------------o
| ( x , y ) |
o-------------------------------------------------o
In traversing the cactus graph, in this case a cactus
of one rooted lobe, one starts at the root, reads off
a left parenthesis "(" on the ascent up the left side
of the lobe, reads off the variable "x", counts off a
comma "," as one transits the interior expanse of the
lobe, reads off the variable "y", and then sounds out
a right parenthesiss ")" on the descent down the last
slope that closes out the clause of this cactus lobe.
According to the current story about how the abstract logical situation
is embodied in the concrete physical situation, the whole pool of units
that corresponds to this expression comes to its resting condition when
just one of the two units in {x, y} is resting and the other is charged.
We may think of the state of the whole pool as associated with the root
node of the cactus, here distinguished by an "amphora" or "at" sign "@",
but the root of the cactus is not represented by an individual agent of
the system, at least, not yet. We may summarize these facts in tabular
form, as shown in Table 5. Simply by way of a common term, let's count
a single unit as a "pool of one".
Table 5. Dynamics of (x , y)
o---------o---------o---------o
| x | y | (x , y) |
o=========o=========o=========o
| charged | charged | charged |
o---------o---------o---------o
| charged | resting | resting |
o---------o---------o---------o
| resting | charged | resting |
o---------o---------o---------o
| resting | resting | charged |
o---------o---------o---------o
I'm going to let that settle a while.
Note 14
Table 5 sums up the facts of the physical situation at equilibrium.
If we let B = {note, rest} = {moving, steady} = {charged, resting},
or whatever candidates you pick for the 2-membered set in question,
the Table shows a function f : B x B -> B, where f[x, y] = (x , y).
Table 5. Dynamics of (x , y)
o---------o---------o---------o
| x | y | (x , y) |
o=========o=========o=========o
| charged | charged | charged |
o---------o---------o---------o
| charged | resting | resting |
o---------o---------o---------o
| resting | charged | resting |
o---------o---------o---------o
| resting | resting | charged |
o---------o---------o---------o
There are two ways that this physical function
might be taken to represent a logical function:
1. If we make the identifications:
charged = true (= indicated),
resting = false (= otherwise),
then the physical function f : B x B -> B
is tantamount to the logical function that
is commonly known as "logical equivalence",
or just plain "equality":
Table 6. Equality Function
o---------o---------o---------o
| x | y | (x , y) |
o=========o=========o=========o
| true | true | true |
o---------o---------o---------o
| true | false | false |
o---------o---------o---------o
| false | true | false |
o---------o---------o---------o
| false | false | true |
o---------o---------o---------o
2. If we make the identifications:
resting = true (= indicated),
charged = false (= otherwise),
then the physical function f : B x B -> B
is tantamount to the logical function that
is commonly known as "logical difference",
or "exclusive disjunction":
Table 7. Difference Function
o---------o---------o---------o
| x | y | (x , y) |
o=========o=========o=========o
| false | false | false |
o---------o---------o---------o
| false | true | true |
o---------o---------o---------o
| true | false | true |
o---------o---------o---------o
| true | true | false |
o---------o---------o---------o
Although the syntax of the cactus language modifies the
syntax of Peirce's graphical formalisms to some extent,
the first interpretation corresponds to what he called
the "entitative graphs" and the second interpretation
corresponds to what he called the "existential graphs".
In working through the present example, I have chosen
the existential interpretation of cactus expressions,
and so the form "(jets , sharks)" is interpreted as
saying that everything in the universe of discourse
is either a Jet or a Shark, but never both at once.
Note 15
Before we tangle with the rest of the Jets and Sharks example,
let's look at a cactus expression that's next in the series
we just considered, this time a lobe with three variables.
For instance, let's analyze the cactus form whose graph
and string expressions are shown in the next display.
o-------------------------------------------------o
| |
| x y z |
| o--o--o |
| \ / |
| \ / |
| @ |
| |
o-------------------------------------------------o
| (x, y, z) |
o-------------------------------------------------o
As always in this competitive paradigm, we assume that
the units x, y, z are mutually inhibitory, so that the
only states that are possible at equilibrium are those
with exactly one unit charged and all the rest at rest.
Table 8 gives the lobal dynamics of the form (x, y, z).
Table 8. Lobal Dynamics of the Form (x, y, z)
o-----------o-----------o-----------o-----------o
| x | y | z | (x, y, z) |
o-----------o-----------o-----------o-----------o
| | | | |
| charged | charged | charged | charged |
| | | | |
| charged | charged | resting | charged |
| | | | |
| charged | resting | charged | charged |
| | | | |
| charged | resting | resting | resting |
| | | | |
| resting | charged | charged | charged |
| | | | |
| resting | charged | resting | resting |
| | | | |
| resting | resting | charged | resting |
| | | | |
| resting | resting | resting | charged |
| | | | |
o-----------o-----------o-----------o-----------o
Given B = {charged, resting} the Table presents the appearance
of a function f : B x B x B -> B, where f[x, y, z] = (x, y, z).
If we make the identifications, charged = false, resting = true,
in accord with the so-called "existential" interpretation, then
the physical function f : B^3 -> B is tantamount to the logical
function that is suggested by the phrase "just 1 of 3 is false".
Table 9 is the truth table for the logical function that we get,
this time using 0 for false and 1 for true in the customary way.
Table 9. Existential Interpretation of (x, y, z)
o-----------o-----------o-----------o-----------o
| x | y | z | (x, y, z) |
o-----------o-----------o-----------o-----------o
| | |
| 0 0 0 | 0 |
| | |
| 0 0 1 | 0 |
| | |
| 0 1 0 | 0 |
| | |
| 0 1 1 | 1 |
| | |
| 1 0 0 | 0 |
| | |
| 1 0 1 | 1 |
| | |
| 1 1 0 | 1 |
| | |
| 1 1 1 | 0 |
| | |
o-----------------------------------o-----------o
Note 16
I sometimes refer to the cactus lobe operators in the series (), (x_1), (x_1, x_2), (x_1, x_2, x_3), ..., (x_1, ..., x_k) as "boundary operators" and one of the reasons for this can be seen most easily in the venn diagram for the k-argument boundary operator (x_1, ..., x_k). Figure 10 shows the venn diagram for the 3-fold boundary form (x, y, z). o-----------------------------------------------------------o | U | | | | o-------------o | | / \ | | / \ | | / \ | | / \ | | / \ | | o o | | | X | | | | | | | | | | | | | | | | | | | o--o----------o o----------o--o | | / \%%%%%%%%%%\ /%%%%%%%%%%/ \ | | / \%%%%%%%%%%o%%%%%%%%%%/ \ | | / \%%%%%%%%/ \%%%%%%%%/ \ | | / \%%%%%%/ \%%%%%%/ \ | | / \%%%%/ \%%%%/ \ | | o o--o-------o--o o | | | |%%%%%%%| | | | | |%%%%%%%| | | | | |%%%%%%%| | | | | Y |%%%%%%%| Z | | | | |%%%%%%%| | | | o o%%%%%%%o o | | \ \%%%%%/ / | | \ \%%%/ / | | \ \%/ / | | \ o / | | \ / \ / | | o-------------o o-------------o | | | | | o-----------------------------------------------------------o Figure 10. Venn Diagram for (x, y, z) In this picture, the "oval" (actually, octangular) regions that are customarily said to be "indicated" by the basic propositions x, y, z : B^3 -> B, that is, where the simple arguments x, y, z, respectively, evaluate to true, are marked with the corresponding capital letters X, Y, Z, respectively. The proposition (x, y, z) comes out true in the region that is shaded with per cent signs. Invoking various idioms of general usage, one may refer to this region as the indicated region, truth set, or fibre of truth of the proposition in question. It is useful to consider the truth set of the proposition (x, y, z) in relation to the logical conjunction xyz of its arguments x, y, z. In relation to the central cell indicated by the conjunction xyz, the region indicated by "(x, y, z)" is composed of the "adjacent" or the "bordering" cells. Thus they are the cells that are just across the boundary of the center cell, arrived at by taking all of Leibniz's "minimal changes" from the given point of departure.
Note 17
Any cell in a venn diagram has a well-defined set of nearest neighbors, and so we can apply a boundary operator of the appropriate rank to the list of signed features that conjoined would indicate the cell in view. For example, having computed the "boundary", or what is more properly called the "point omitted neighborhood" (PON) of the center cell in a 3-dimensional universe of discourse, what is the PON of the cell that is furthest from it, namely, the "origin cell" indicated as (x)(y)(z)? The region bordering the origin cell, (x)(y)(z), can be computed by placing its three signed conjuncts in a 3-place bracket like (__, __, __), arriving at the cactus expression that is shown below in both graph and string forms. o-------------------------------------------------o | | | x y z | | o o o | | | | | | | o--o--o | | \ / | | \ / | | @ | | | o-------------------------------------------------o | ((x),(y),(z)) | o-------------------------------------------------o Figure 11 shows the venn diagram of this expression, whose meaning is adequately suggested by the phrase "just 1 of 3 is true". o-----------------------------------------------------------o | U | | | | o-------------o | | /```````````````\ | | /`````````````````\ | | /```````````````````\ | | /`````````````````````\ | | /```````````````````````\ | | o`````````````````````````o | | |``````````` X ```````````| | | |`````````````````````````| | | |`````````````````````````| | | |`````````````````````````| | | |`````````````````````````| | | o--o----------o```o----------o--o | | /````\ \`/ /````\ | | /``````\ o /``````\ | | /````````\ / \ /````````\ | | /``````````\ / \ /``````````\ | | /````````````\ / \ /````````````\ | | o``````````````o--o-------o--o``````````````o | | |`````````````````| |`````````````````| | | |`````````````````| |`````````````````| | | |`````````````````| |`````````````````| | | |``````` Y ```````| |`````` Z ````````| | | |`````````````````| |`````````````````| | | o`````````````````o o`````````````````o | | \`````````````````\ /`````````````````/ | | \`````````````````\ /`````````````````/ | | \`````````````````\ /`````````````````/ | | \`````````````````o`````````````````/ | | \```````````````/ \```````````````/ | | o-------------o o-------------o | | | | | o-----------------------------------------------------------o Figure 11. Venn Diagram for ((x),(y),(z))
Note 18
Given the foregoing explanation of the k-fold boundary operator,
along with its use to express such forms of logical constraints
as "just 1 of k is false" and "just 1 of k is true", there will
be no trouble interpreting an expression of the following shape
from the Jets and Sharks example:
(( art ),( al ),( sam ),( clyde ),( mike ),
( jim ),( greg ),( john ),( doug ),( lance ),
( george ),( pete ),( fred ),( gene ),( ralph ),
( phil ),( ike ),( nick ),( don ),( ned ),
( karl ),( ken ),( earl ),( rick ),( ol ),
( neal ),( dave ))
This expression says that everything in the universe of discourse
is either Art, or Al, or ..., or Neal, or Dave, but never any two
of them at once. In effect, I've exploited the circumstance that
the universe contains but finitely many ostensible individuals to
dedicate its own predicate to each one of them, imposing only the
requirement that these predicates must be disjoint and exhaustive.
Likewise, each of the following clauses has the effect of
partitioning the universe of discourse among the factions
or features that are enumerated in the clause in question.
( jets , sharks )
(( 20's ),( 30's ),( 40's ))
(( junior_high ),( high_school ),( college ))
(( single ),( married ),( divorced ))
(( bookie ),( burglar ),( pusher ))
We may note in passing that ( x , y ) = ((x),(y)),
but a rule of this form holds only in the case of
the 2-fold boundary operator.
Note 19
Let's collect the various ways of representing the structure of a universe of discourse that is described by the following cactus expressions, verbalized as "just 1 of x, y, z is true". o-------------------------------------------------o | | | x y z | | o o o | | | | | | | o--o--o | | \ / | | \ / | | @ | | | o-------------------------------------------------o | ((x),(y),(z)) | o-------------------------------------------------o Table 12 shows the truth table for the existential interpretation of the cactus formula ((x),(y),(z)). Table 12. Existential Interpretation of ((x),(y),(z)) o-----------o-----------o-----------o-------------o | x | y | z | (x, y, z) | o-----------o-----------o-----------o-------------o | | | | 0 0 0 | 0 | | | | | 0 0 1 | 1 | | | | | 0 1 0 | 1 | | | | | 0 1 1 | 0 | | | | | 1 0 0 | 1 | | | | | 1 0 1 | 0 | | | | | 1 1 0 | 0 | | | | | 1 1 1 | 0 | | | | o-----------------------------------o-------------o Figure 13 shows the same data as a 2-colored 3-cube, coloring a node with a hollow dot (o) for "false" or a star (*) for "true". o-------------------------------------------------o | | | x y z | | o | | /|\ | | / | \ | | / | \ | | / | \ | | / | \ | | / | \ | | / x (y) z \ | | x y (z) o o o (x) y z | | |\ / \ /| | | | \ / \ / | | | | \ / \ / | | | | \ / | | | | / \ / \ | | | | / \ / \ | | | |/ \ / \| | | x (y)(z) * * * (x)(y) z | | \ (x) y (z) / | | \ | / | | \ | / | | \ | / | | \ | / | | \ | / | | \|/ | | o | | (x)(y)(z) | | | o-------------------------------------------------o Figure 14 repeats the venn diagram that we've already seen. o-----------------------------------------------------------o | U | | | | o-------------o | | /```````````````\ | | /`````````````````\ | | /```````````````````\ | | /`````````````````````\ | | /```````````````````````\ | | o`````````````````````````o | | |``````````` X ```````````| | | |`````````````````````````| | | |`````````````````````````| | | |`````````````````````````| | | |`````````````````````````| | | o--o----------o```o----------o--o | | /````\ \`/ /````\ | | /``````\ o /``````\ | | /````````\ / \ /````````\ | | /``````````\ / \ /``````````\ | | /````````````\ / \ /````````````\ | | o``````````````o--o-------o--o``````````````o | | |`````````````````| |`````````````````| | | |`````````````````| |`````````````````| | | |`````````````````| |`````````````````| | | |``````` Y ```````| |`````` Z ````````| | | |`````````````````| |`````````````````| | | o`````````````````o o`````````````````o | | \`````````````````\ /`````````````````/ | | \`````````````````\ /`````````````````/ | | \`````````````````\ /`````````````````/ | | \`````````````````o`````````````````/ | | \```````````````/ \```````````````/ | | o-------------o o-------------o | | | | | o-----------------------------------------------------------o Figure 14. Venn Diagram for ((x),(y),(z)) Figure 15 shows an alternate form of venn diagram for the same proposition, where we collapse to a nullity all of the regions on which the proposition in question evaluates to false. This leaves a structure that partitions the universe into precisely three parts. In mathematics, operations that identify diverse elements are called "quotient operations". In this case, many regions of the universe are being identified with the null set, leaving only this 3-fold partition as the "quotient structure". o-----------------------------------------------------------o | \ / | | \ / | | \ / | | \ / | | \ / | | \ X / | | \ / | | \ / | | \ / | | \ / | | \ / | | \ / | | \ / | | \ / | | o | | | | | | | | | | | | | | Y | Z | | | | | | | | | | | | | | | | | | | | | | | | | | | | o-----------------------------o-----------------------------o Figure 15. Quotient Structure Venn Diagram for ((x),(y),(z))
Note 20
Let's now look at the last type of clause that we find in my
transcription of the Jets and Sharks data base, for instance,
as exemplified by the following couple of lobal expressions:
( jets ,
( art ),( al ),( sam ),( clyde ),( mike ),
( jim ),( greg ),( john ),( doug ),( lance ),
( george ),( pete ),( fred ),( gene ),( ralph ))
( sharks ,
( phil ),( ike ),( nick ),( don ),( ned ),( karl ),
( ken ),( earl ),( rick ),( ol ),( neal ),( dave ))
Each of these clauses exhibits a generic pattern whose logical properties
may be studied well enough in the form of the following schematic example.
o-------------------------------------------------o
| |
| y z |
| o o |
| x | | |
| o--o--o |
| \ / |
| \ / |
| @ |
| |
o-------------------------------------------------o
| ( x ,(y),(z)) |
o-------------------------------------------------o
The proposition (u, v, w) evaluates to true
if and only if just one of u, v, w is false.
In the same way, the proposition (x,(y),(z))
evaluates to true if and only if exactly one
of x, (y), (z) is false. Taking it by cases,
let us first suppose that x is true. Then it
has to be that just one of (y) or (z) is false,
which is tantamount to the proposition ((y),(z)),
which is equivalent to the proposition ( y , z ).
On the other hand, let us suppose that x is the
false one. Then both (y) and (z) must be true,
which is to say that y is false and z is false.
What we have just said here is that the region
where x is true is partitioned into the regions
where y and z are true, respectively, while the
region where x is false has both y and z false.
In other words, we have a "pie-chart" structure,
where the genus X is divided into the disjoint
and X-haustive couple of species Y and Z.
The same analysis applies to the generic form
(x, (x_1), ..., (x_k)), specifying a pie-chart
with a genus X and the k species X_1, ..., X_k.
Differential Logic : Series C
It would be good to summarize, in rough but intuitive terms, the outlook on differential logic that we have reached so far.
We have been considering a class of operators on universes of discourse, each of which takes us from considering one universe of discourse, X, to considering a larger universe of discourse, EX.
Each of these operators, in general terms having the form F : X -> EX, acts on each proposition p : X -> B of the source universe X to produce a proposition Fp : EX -> B of the target universe EX.
The two main operators that we have worked with up to this point are the enlargement operator E : X -> EX and the difference operator D : X -> EX.
E and D take a proposition in X, that is, a proposition p : X -> B that is said to be "about" the subject matter of X, and produce the extended propositions Ep, Dp : EX -> B, which may be interpreted as being about specified collections of changes that might occur in X.
Here we have need of visual representations, some array of concrete pictures to anchor our more earthy intuitions and to help us keep our wits about us before we try to climb any higher into the ever more rarefied air of abstractions.
One good picture comes to us by way of the "field" concept. Given a space X, a "field" of a specified type T over X is formed by assigning to each point of X an object of type T. If that sounds like the same thing as a function from X to the space of things of type T, it is, but it does seems to help to vary the mental pictures and the figures of speech that naturally spring to mind within these fertile fields.
In the field picture, a proposition p : X -> B becomes a "scalar" field, that is, a field of values in B, or a "field of true-false indications".
Let us take a moment to view an old proposition in this new light, for example, the conjunction uv : X -> B that is depicted in Figure 1.
o-------------------------------------------------o | X | | | | o-------------o o-------------o | | / \ / \ | | / o \ | | / /`\ \ | | / /```\ \ | | o o`````o o | | | |`````| | | | | U |`````| V | | | | |`````| | | | o o`````o o | | \ \```/ / | | \ \`/ / | | \ o / | | \ / \ / | | o-------------o o-------------o | | | | | o-------------------------------------------------o | f = u v | o-------------------------------------------------o Figure 1. Conjunction uv : X -> B
Each of the operators E, D : X -> EX takes us from considering propositions p : X -> B, here viewed as "scalar fields" over X, to considering the corresponding "differential fields" over X, analogous to what are usually called "vector fields" over X.
The structure of these differential fields can be described this way. To each point of X there is attached an object of the following type, a proposition about changes in X, that is, a proposition g : dX -> B. In this setting, if X is the universe that is generated by the set of coordinate propositions {u, v}, then dX is the differential universe that is generated by the set of differential propositions {du, dv}. These differential propositions may be interpreted as indicating "change in u" and "change in v", respectively.
A differential operator F, of the first order sort that we have been considering, takes a proposition p : X -> B and gives back a differential proposition Fp : EX -> B.
In the field view, we see the proposition p : X -> B as a scalar field and we see the differential proposition Fp : EX -> B as a vector field, specifically, a field of propositions about contemplated changes in X.
The field of changes produced by E on uv is shown in Figure 2.
o-------------------------------------------------o | X | | | | o-------------o o-------------o | | / \ / \ | | / U o V \ | | / /`\ \ | | / /```\ \ | | o o.->-.o o | | | u(v)(du)dv |`\`/`| (u)v du(dv) | | | | o---------------|->o<-|---------------o | | | | |``^``| | | | o o``|``o o | | \ \`|`/ / | | \ \|/ / | | \ o / | | \ /|\ / | | o-------------o | o-------------o | | | | | | | | | | | o | | (u)(v) du dv | | | o-------------------------------------------------o | f = u v | o-------------------------------------------------o | | | Ef = u v (du)(dv) | | | | + u (v) (du) dv | | | | + (u) v du (dv) | | | | + (u)(v) du dv | | | o-------------------------------------------------o Figure 2. Enlargement E[uv] : EX -> B
The differential field E[uv] specifies the changes that need to be made from each point of X in order to reach one of the models of the proposition uv, that is, in order to satisfy the proposition uv.
The field of changes produced by D on uv is shown in Figure 3.
o-------------------------------------------------o | X | | | | o-------------o o-------------o | | / \ / \ | | / U o V \ | | / /`\ \ | | / /```\ \ | | o o`````o o | | | (du)dv |`````| du(dv) | | | | o<--------------|->o<-|-------------->o | | | | |``^``| | | | o o``|``o o | | \ \`|`/ / | | \ \|/ / | | \ o / | | \ /|\ / | | o-------------o | o-------------o | | | | | | | | v | | o | | du dv | | | o-------------------------------------------------o | f = u v | o-------------------------------------------------o | | | Df = u v ((du)(dv)) | | | | + u (v) (du) dv | | | | + (u) v du (dv) | | | | + (u)(v) du dv | | | o-------------------------------------------------o Figure 3. Difference D[uv] : EX -> B
The differential field D[uv] specifies the changes that need to be made from each point of X in order to change the value of the proposition uv.
References
- Edelman, Gerald M. (1988), Topobiology : An Introduction to Molecular Embryology, Basic Books, New York, NY.
