1

I am trying to understand Gödel's first incompleteness theorem from his original 1931 paper.

Here is a translation i am using for my studies :

http://www.research.ibm.com/people/h/hirzel/papers/canon00-goedel.pdf

I feel like i now have a decent intuitive understanding of it all.

Now I'm trying to understand all the technical details of it.

I don't fully understand these so called "elementary-formulae", and how it makes sense of my intuitive understanding of what they are :

For example, i suppose "$y = x+1$" is an elementary formula, how does that fit that "elementary-formulae" a(b) with 'a' a type-n+1 variable and 'b' a type-n variable.

I figure "$y = x+1$" would be "captured" by the type-3 variable :

{ (0,1), (1,2), ..., (n,n+1) }

= { {{0},{0,1}}, {{1},{1,2}}, ..., {{n},{n,n+1}} }

how would i write it as :

"'type-n+1 variable' ( 'type-n variable' ) " ?

Or can someone give me some examples of elementary formulae in the Gödel context ?

Thank you.

1 Answers1

2

Unfortunately, the formal language of Gödel's paper is today quite unused :

By a sign of type $1$ we understand a combination of signs that has [anyone of] the forms

$$a, fa, ffa, fffa, \ldots,$$

and so on, where $a$ is either $0$ or a variable of type $1$. In the first case, we call such a sign a numeral. [...] A combination of signs that has the form $a(b)$, where $b$ is a sign of type $n$ and a a sign of type $n + 1$, will be called an elementary formula.

Thus [see note 21 of standard translation] "$=$" is not a primitive sign, but is defined with quantification on second-order (unary) predicate variables. The simple formula :

$$x_1=y_1+1$$

in Gödel's system is :

$$x_2 \Pi (x_2(x_1) \to x_2(fy_1))$$

and thus it is not an elementary one.

As you can see, Gödel introduces only (unary) predicates :

Remark: Variables for functions of two or more argument places (relations) need not be included among the primitive signs since we can define relations to be classes of ordered pairs, and ordered pairs to be classes of classes; for example, the ordered pair $a, b$ can be defined to be $((a), (a, b))$, where $(x, y)$ denotes the class whose sole elements are $x$ and $y$, and $(x)$ the class whose sole element is $x$.

Thus, the only elementary formulae are like : $x_2(x_1)$ or $x_2(f0)$.


If we restrict ourselves to the more usual first-order language of arithmetic, the only elementary formulae are expressions like :

$t=s$

where $t,s$ are terms, i.e. (individual) variables, the constant $0$ or "complex" terms like the numerals: $ff \ldots f0$.

To be formal, instead of e.g. $x=f0$, we have to write :

$=(x,f0)$.

Thus, taking into account that :

  1. $E(x) = R(11)*x*R(13)$,

i.e. $E(x)$ corresponds to the operation of "enclosing within parentheses", we have to modify the definition :

  1. $Elf(x) = (\exists y, z, n)[y, z, n \le x \land Typ_n(y) \land Typ_{n+1}(z) \land x = z*E(y)]$,

i.e. the relation encoding "$x$ is an ELEMENTARY FORMULA", in order to take into account binary relations.

In Peter Smith's book, Ch.20 Arithmetization in more details, page 144-on, you can find the def for the first-order language with "$=$" as primitive :

R13. The property $Atom(n)$ which holds when $n$ is the g.n. of an atomic wff is primitive recursive.

$Atom(n) \overset{def}{=} (∃x ≤ n)(∃y ≤ n)[Term(x) ∧ Term(y) ∧ n = (x* \ulcorner = \urcorner *y)]$.

Thus, in our "simplified" example: $=(x,f0)$, we have that $=$ is a sign of Type $2$ (it is a predicate) while $x$ (a variable) and $f0$ (a numeral) are terms, i.e. signs of Type 1.

  • thank you for your answer. Let me rephrase : if i'm correct, an elementary formulae is a(b) with 'a' being the '=' sign, hence (i guess) the set of sets { {1,1}, ... , {n,n}, ... } a type 3 variable and 'b' being ... { {x}, {x, f0} } a type 2 varible, or {x, f0}, a type 2 variable ? – joseph M'Bimbi-Bene Oct 29 '15 at 21:02
  • @MauroALLEGRANZA: Gödel's paper deals with the system of Russell and Whitehead's *Principia Mathematica", not some system of his own devising as your answer suggests. – Rob Arthan Nov 01 '15 at 21:33
  • @RobArthan - not exactly; see van Heijennort, page 595 (first para of G's paper) : "If to the Peano axioms we add the logic of Principia mathematica (with the natural numbers as the individuals) together with the axiom of choice (for all types), we obtain a formal system $S$, for which the following theorems hold ...". For the details of arithmetization (at the level required by the OP) we have to stay closer to G's system. – Mauro ALLEGRANZA Nov 02 '15 at 08:18
  • Thank you for your reply and please excuse me for answering late.

    The thing is that i don't understand the PM notation and logic. I have the 1962 translation but it doesn't explain much.

    It is hard to get a hand on a copy of principia mathematica and even then, it's not very accessible.

    I still managed to find it there : http://quod.lib.umich.edu/cgi/t/text/text-idx?c=umhistmath;idno=AAT3201.0001.001

    I don't know exactly how to make sense of x2Π(x2(x1)→x2(fy1)) . you defined x2 as an unary predicate sign

    – joseph M'Bimbi-Bene Nov 06 '15 at 10:55
  • you said in one of my other post that you didn't understand my "set of set thing", but it's Gödel himself who talks about it :

    Variable of type one (for individuals, i.e. natural numbers including 0):

    Variables of type two (for classes of individuals , i.e. subsets of IN ):

    Variables of type three (for classes of classes of individuals , i.e. sets of subsets of N ):

    And so on for every natural number as type.

    We call combinations of signs of the form a(b), where b is a sign of type n and a a sign of type n + 1, elementary formulae.

    – joseph M'Bimbi-Bene Nov 06 '15 at 10:59
  • so according to his scheme, for an arbitrary "elementary formula" x2(x1), x2 has to be an (n+1)-order set of natural number, and x1 has to be an (n)-order set of natural number

    That's why i bring that "set of sets" thing.

    – joseph M'Bimbi-Bene Nov 06 '15 at 11:01
  • and "x2Π(x2(x1)→x2(fy1))" seems unidirectionnal to me. shouldn't it be something like "x2Π(x2(x1)→x2(fy1)) ∧ x2Π(x2(fy1))→ x2(x1)", from what i understand about implication and equivalence in logic ? – joseph M'Bimbi-Bene Nov 06 '15 at 11:02
  • @josephM'Bimbi-Bene - regarding $x_2Π(x_2(x_1) → x_2(fy_1))$ it is simply (in "modern" notation : $\forall P[P(x_1) → P(f(y_1)))]$, where $f$ is the successor function. – Mauro ALLEGRANZA Nov 06 '15 at 12:27
  • Regarding "set of sets", the answer is : NO. $x_2$ is a sign (i.e. an expression of the language). In a "modern" exposition of (high-order) logic, is a (unary) predicate variable, like : $P_1, P_2,\ldots$. In order to "let the machine works" (I'll referrring to coding and so on) it is not necessarily to interpret them. – Mauro ALLEGRANZA Nov 06 '15 at 12:29
  • I have a feeling this "set of sets of sets of natural number" things is just the so-called extension of either a function or a relation.

    I eventually also reached the conclusion that "x2(x1)" would be written in a modern notation as "P(x1)" but that somehow, x2 would not be represented by a "formula" (with '+', 'x' etc.) but by its extension in his system and x2(x1) is just a "shortand" sort of representation ... am i correct ?

    – joseph M'Bimbi-Bene Nov 06 '15 at 14:26
  • so x2 is a sign for a "property"/"class" of whatever you call it, whose extension is a set of sets of sets of natural number.

    But we don't use its extension to represent that property. Now my question whould be : "how do we represent, by a "formula" a property ?" In his formalism, it seems like it's not possible.

    And are there sets (of sets of sets) who cannot be represented by a formula.

    I think i read somewhere something about Frege stating something relative to it

    – joseph M'Bimbi-Bene Nov 06 '15 at 14:29
  • And regarding $$x_2 \Pi (x_2(x_1) \to x_2(fy_1))$$, i understand how it's supposed to be formulated in a modern fashion but i still can't make sense of it.

    How does that make sense that a property on some element 'a' implying the same property on the element 'b' make 'a' and 'b' equal ?

    – joseph M'Bimbi-Bene Nov 06 '15 at 14:33
  • @josephM'Bimbi-Bene - Yes; $x_2$ is a sign for a "set" or "class" of numbers. In G's paper : "Variables of type $1$ (for individuals, that is, natural numbers including $0$) : "$x_1$" [...] Variables of type $2$ (for classes of individuals): "$x_2". – Mauro ALLEGRANZA Nov 06 '15 at 14:36
  • How to express a property of numbers in G's formalism ? Assuming the "defined" symbol "$=$" (that we know how to "unwind" in G's primitive formalism) the following is a property : $Pos(x_1) \leftrightarrow \lnot(x_1 = 0)$. As you can see, $Pos(x_1)$ holds iff $x_1$ is "positive" natural number, i.e. $x_1 > 0$. – Mauro ALLEGRANZA Nov 06 '15 at 14:38
  • Regarding G's formula for $=$, it is simply the (standard) second-order definition of identity due to Leibniz (see Identity of indiscernibles) and adopted by Whitehead & Russell in PM. In words : "two "individuals" $x$ and $y$ are identical if every predicate possessed by $x$ is also possessed by $y$ and vice versa". I've applied it to the individuals (i.e. numbers : $x_1$ and $fx_1$. – Mauro ALLEGRANZA Nov 06 '15 at 14:43
  • thank you a lot for your link => "identity of indiscernibles". But still it is defined with an equivalence, not an implication (i also found an excerpt of Principia with the definition of the equality, i also seems to use the implication. Maybe in that special case both strictly equivalent ...) – joseph M'Bimbi-Bene Nov 07 '15 at 09:15
  • i will work and read on that a little bit more but it's more clear now. Thank you – joseph M'Bimbi-Bene Nov 07 '15 at 09:17
  • Hello, A year later, looking back at it, I am still stuck, i still don't have a solid understanding of what an elementary formula is :( .

    Could it be possible to provide me with a trivial step by step example of one such elemenary formula, with constants and variables, and constructions of formulae with 'or' and 'forall' operators so i can figure that all out ... thank you in advance if you could do that (since you seem to be very knowledgeable about Gödel and (oldstyle) mathematical logic overall

    Thank you

    – joseph M'Bimbi-Bene Nov 03 '16 at 09:05
  • OK from the formula x2Π(x2(x1)→x2(fy1)) i get the intuition of formulae using 'not', 'or' and the universal quantifier. But I really have a hard time understanding elementary formulae from type-n / type-n+1 variables and classes of [...] classes of natural numbers, i really have a hard time working my head around that.

    If you could help me on that, i would be sooo grateful (hell i would even give you money if i could ^^ !!)

    – joseph M'Bimbi-Bene Nov 03 '16 at 09:38
  • @josephM'Bimbi-Bene - You have to "prcatice" ... Unfortunately, G's formalism is not user-friendly, especially if you are accustomed with "modern" symbols. Try with a simple example in semi-formal language: we write $Red(x)$ where $x$ is a type-1 symbol, i.e. stand for an object. Thus $\forall x Red(x)$ means the (false) statement that "every object is red". Consider now the type-1 symbol $f0$ (the "name" for the number $1$) and consider a type-2 symbol $P$ i.e. a symbol that stand for a property (an entiry of a higher level with respect to "objects"). 1/n – Mauro ALLEGRANZA Nov 03 '16 at 10:02
  • The formula $\forall P P(f0)$ means the (false) statement that "the number $1$ has every properties". This fact is carachteristic of high-order logic and this is the second complex feature of G's paper (compared to modern treatments like Smith's book, that used first-order logic). 2/n – Mauro ALLEGRANZA Nov 03 '16 at 10:03
  • In f-o logic equality is a primitive symbol and needs specific axioms. In high-order logic equality is defined trough the formula : $x=y \leftrightarrow \forall P [P(x) \to P(y)]$. The intuitive meaning is : "two objects are equal (or identical) iff they have the same properties, i.e. if we cannot find a property that discriminate between them". See Identity of Indiscernibles (due to Leibniz). 3/n – Mauro ALLEGRANZA Nov 03 '16 at 10:07
  • Thus, if we have to rewrite the above principle into G's system. It runs "for every property $P$, if $x$ has $P$, then also $y+1$ has $P$". Re-read it as : "for every type-2 symbol $x_2$ , if the type-1 symbol $x_1$ has $x_2$, then also the type-1 symbol $y_1+1$ has it". In symbols : $x_2 \Pi (x_2(x_1) \to x_2(fy_1))$. last/n – Mauro ALLEGRANZA Nov 03 '16 at 10:12
  • Thus, if we restrict ourselves to the "bottom levels" of the hierarchy, we have the level-0 made of individual objects (denoted by type-1 symbols) and the level-1 made of collections of objects (denoted by type-2 symbols), and so on. – Mauro ALLEGRANZA Nov 03 '16 at 11:30
  • @MauroALLEGRANZA I have a similar question so I thought to post a comment here. How do you define addition in Gödel's system? The proof of proposition V requires addition and multiplication to be definable of course, but it doesn't seem easy with no functional symbols. Maybe something obvious is missed. – Tony Jun 23 '19 at 07:31