Skip to main content

Gatterdam’s proof that free products with amalgamation are En+1-computable

I haven’t posted any research in ages! I haven’t done any research in ages! Here’s a rewriting of a theorem from Ronald Gatterdam’s doctoral thesis, in more sensible language. I’ve proved the same theorem for fundamental groups of graphs of groups.

The main problem with Gatterdam’s proof was that he didn’t invent any new notation to make life simpler, so he wrote out pages and pages of very repetitive function applications.

I’ve made up a lot of notation to make dealing with functions in the Grzegorczyk hierarchy a lot easier, so you’ll probably need to refer to my draft paper for a few things. I keep all my research notes in a repository on github, by the way.

Theorem

Let G1,G2 be En-computable groups. Let H1,H2 be En-decidable subgroups of the latter. Let ϕ:H1H2 be an isomorphism, with ϕ,ϕ1 both En-computable.

Then the free product of G1 and G2 with H1 and H2 amalgamated, G=G1ϕG2, is En+1-computable.

Proof

Let (i1,m1,j1) be the index of G1 and (i2,m2,j2) be the index of G2. The dashes will be explained later.

We can assume, without loss of generality, that 0ia(Ga) and ia(1)=1 for a=1,2. (Might not even need this.)

By Magnus, Karrass, Solitar, etc., all elements gG have normal form

g=hg1gr

where hH1, and the gi are coset representatives of Ga/Ha, a=1,2, such that gi+1G1giG2.

The following proof becomes a lot easier if we redefine the factor group indices as follows:

i1(x):=2i1(x),m1(x,y):=2m1(x2,y2)j1(x):=2j1(x2).

i2(x):=2i2(x)1,m2(x,y):=2m2(x+12,y+12)j2(x):=2j2(x+12).

Now,

xi1(G1)2xx2i1(G1),xi1(H1)xi1(G1)x2i1(H1),xi2(G2)2xx+12i2(G2),xi2(H2)xi2(G2)x+12i2(H2).

The subgroup isomorphism also needs to be redefined:

ϕ(x):=2ϕ(x2)1ϕ1(x):=2ϕ1(x+12)

In order to do multiplication, we need to be able to split every gaGa, a=1,2, into a word of the form haka, where haHa and ka is a coset representative of ga in Ga/Ha.

Define:

ka(x):=minyx(ma(x,ja(y))ia(Ha)),ha(x):=ma(x,ja(ka(x))).

Now we can define i(G) by

i(hg1gr):=[h,g1,,gr].

(encode it as a Gödel number)

i(G) is En-decidable because xi(G) if and only if:

  • x is a Gödel number
  • (x)0i1(H1)
  • 2<i<|x|, (x)ii1(G1)(x)i1i2(G2)
  • 1<i<|x|, (x)iia(Ga)ka((x)i)=(x)i.

Now define a function r:(i1(G1)i2(G2))×i(G)i(G) which does multiplication of elements of G on the left by elements of G1 and G2.

r is defined by cases in a decision tree, so here’s some programming-ish syntax:

r(x,y):=

If {xi1(G1)}

If {xi1(H1)}

[m1(x,(y)0)]++y[1]

else

If {(y)1i1(G1)}

If {m1(m1(x,(y)0),(y)1)i1(H1) }

[m1(m1(x,(y)0),(y)1)]++y[2]

else

[h1m1(m1(x,(y)0),(y)1),k1m1(m1(x,(y)0),(y)1)]++y[2]

endif

else

[h1m1(x,(y)0),k1m2(x,(y)0)]++y[1]

endif

endif

else

If {xi2(H2) }

[m1(ϕ1(x),(y)0)]++y[1]

else

If {(y)1i2(G2) }

If {m2(m2(x,ϕ((y)0)),(y)1)i2(H2) }

[ϕ1m2(m2(x,ϕ((y)0)),(y)1)]++y[2]

else

[ϕ1h2m2(m2(x,ϕ((y)0)),(y)1),k2m2(m2(x,ϕ((y)0)),(y)1)]++y[2]

endif

else

[ϕ1h2m2(x,ϕ((y)0)),k2m2(x,ϕ((y)0))]++y[1]

endif

endif

endif

 

Now we can do multiplication on G in general:

m([],y):=y,m(x:xs,y):=r(x,m(xs,y)).

And inversion:

j([1]):=[1],j(x:xs):=r(j^(x),j(xs)).

j^(x):={j1(x),xi1(G1),j2(x),xi2(G2).

(x:xs means a list of the form [x,x1,x2,]. I borrowed this syntax from Haskell. I need to show that you can define recursive functions on lists like this, but I hope you agree it works)

The functions m and j are defined recursively. Because we don’t know anything about how quickly the original multiplication functions grow, the recursion is unbounded, so the best we can do is to say that the index of G is at worst En+1-computable.

QED

Gatterdam also shows that the embeddings GaG are natural, but I lost the will.

Comments

Comments

‘Free product with amalgamation’, also known an a pushout of groups. 🙂 This result is really quite useful for computing homotopy groups using the van Kampen theorem, which states that given a pushout of pointed spaces space with all maps open embeddings, and all spaces path connected, then the fundamental group of the pushout is the pushout of the fundamental groups. I have no idea what ‘E^n computable’ means, but I’m sure it is nice!