Stanley’s theory of P-partitions and the Hook Formula
Wildon's Weblog 2020-03-22
The aim of this post is to introduce Stanley’s theory of labelled -partitions and, as an application, give a short motivated proof of the Hook Formula for the number of standard Young tableaux of a given shape. In fact we prove the stronger -analogue, in which hook lengths are replaced with quantum integers. All the ideas may be found in Chapter 7 of Stanley’s book Enumerative Combinatorics II so, in a particularly strong sense, no originality is claimed.
The division into four parts below and the recaps at the start of each part have the aim of reducing notational overload (the main difficulty in the early parts), while also giving convenient places to take a break.
Part 1: Background on arithmetic partitions
Recall that an arithmetic partition of size is a weakly decreasing sequence of natural numbers whose sum is . For example, there are partitions of with at most three parts, namely
as represented by the Young diagrams shown below.
Since the Young diagram of a partition into at most parts is uniquely determined by its number of columns of lengths , and , such partitions are enumerated by the generating function
For example contributes to the coefficient of , obtained when we expand the geometric series by choosing from
then from
and finally from
Now suppose that we are only interested in partitions where the first part is strictly bigger than the second. Then the Young diagram must have a column of size , and so we replace in the generating function with . Since the coefficient of above is , it follows (without direct enumeration) that there are such partitions. What if the second part must also be strictly bigger than the third? Then the Young diagram must have a column of size , and so we also replace with . (I like to see this by mentally removing the first row: the remaining diagram then has a column of size , by the case just seen.) By a routine generalization we get the following result: partitions with at most parts such that the th largest part is strictly more than the th largest part for all are enumerated by
Part 2: -partitions
Let be a poset with partial order . A -partition is an order-preserving function .
For example, if with the usual order then is the sequence of values of a -partition if and only if . Thus, by removing any initial zeros and reversing the sequence, -partitions are in bijection with arithmetic partitions having at most parts.
I should mention that it is more usual to write rather than . More importantly, Stanley’s original definition has ‘order-preserving’ rather than ‘order-reversing’. This fits better with arithmetic partitions, and plane-partitions, but since our intended application is to reverse plane-partitions and semistandard Young tableaux, the definition as given (and used for instance in Stembridge’s generalization) is most convenient.
Reversed plane partitions
Formally the Young diagram of the partition is the set
of boxes. We partially order the boxes by (the transitive closure of) and . This is shown diagramatically below.
For this partial order, -partitions correspond to assignments of non-negative integers to such that that the rows and columns are weakly increasing when read left to right and top to bottom. For example, three of the -partitions of size are shown below
Such assignments are known as reverse plane partitions. The proof of the Hook Formula given below depends on finding the generating function for reverse plane partitions in two different ways: first using the general theory of -partitions, and then in a more direct way, for instance using the Hillman–Grassl bijection.
Enumerating
As a warm-up, we replace with the smaller partition , so now , ordered by , . Comparing and , we divide the -partitions into two disjoint classes: those with , and those with . The first class satisfy
and so are in bijection with arithmetic partitions with at most parts. The second class satisfy
so are in bijection with arithmetic partitions with at most parts, whose largest part is strictly the largest. By the first section we deduce that
The cancellation to leave a unit numerator shows one feature of the remarkably nice generating function for reverse plane partitions revealed in the final part below.
Labelled -partitions
A surprisingly helpful way to keep track of the weak/strong inequalities seen above is to label the poset elements by natural numbers. We define a labelling of a poset of size to be a bijective function . Suppose that covers in the order on . Either , in which case we say the labelling is natural for , or , in which case we say the labelling is strict for . A -partition is then an order preserving function such that if is a covering relation and
- (natural) then ;
- (strict) then .
Note that the role of the labelling is only to distinguish weak/strong inequalities: the poset itself determines whether or for each comparable pair , .
Let denote the set of -partitions. For example, if as in the example above, then the -partitions are precisely the -partitions for any all-natural labelling; the two choices are
and
Working with , the partitions with form the set where is the total order refining such that
and the partitions with form the set where
The division of -partitions above is an instance of the following result.
Fundamental Lemma. Let be a poset with partial order and let be a labelling. Then
where the union over all total orders refining is disjoint.
Proof. Every partition appears in the right-hand side for some : just choose so that if then and if and then . On the other hand, suppose that is in both and . Choose and incomparable under and such that and . From we get and from we get . Therefore . Now using the labelling for the first time, we may suppose without loss of generality that , and using that and we get , a contradiction. .
Suggested exercise.
Show that the generating functions enumerating and are and
there are respectively and linear extensions that must be considered.
Part 3: Permutations
Recall that is a poset, is a bijective labelling and that a -partition is a function such that if then , with strict inequality whenever .
Connection with permutations
We write permutations of in one-line form as . Recall that has a descent in position if .
Example. Let , be the partial order used above to enumerate , as labelled by , , . The total order corresponds under to the identity permutation of , with no descents. The total order corresponds under to the permutation swapping and , with descent set .
In general, let be a total order refining . Let and consider the elements and of labelled and . In any -partition we have , with strict inequality if and only if . Therefore, using the total order to identify with a function on , i.e. , we require for all , with strict inequality if and only if . Equivalently,
with strict inequality whenever there is a descent in the permutation corresponding under to . Conversely, a permutation of corresponds to a total order refining if and only if appears left of whenever . Therefore Stanley’s Fundamental Lemma may be restated as follows.
Fundamental Lemma restated. Let be a poset with partial order and let be a labelling such that if then appears before in . Then, using the labelling to identity elements of with functions on ,
where is the set of all functions such that with strict inequality whenever , and the union is disjoint.
Such sequences correspond to partitions whose th largest part is strictly more than their th largest part (which might be zero), and so are enumerated by
The power of in the numerator is, by the standard definition, the comajor index of the permutation .
Example
We enumerate . There are extensions of the partial order on ,
corresponding under the labelling
to the permutations with descent sets , and comajor indices , , , , , respectively. By the restatement of the fundamental lemma and the following remark,
We end this part by digressing to outline a remarkably short proof of an identity due to MacMahon enumerating permutations by major index and descent count.
Exercise
If and all elements are incomparable under , then a -partition is simply a function .
- How are such partitions enumerated by the restated Fundamental Lemma?
- Deduce that
where the sum is over all permutations of .
- Give an involution on permutations that preserves the number of descents and swaps the comajor and major indices. Deduce that can be replaced with above.
Exercise
The argument at the start of this post shows that enumerates partitions with at most parts by their size (power of ) and largest part (power of ).
- Show that partitions whose th largest part is strictly larger than their th largest part for all are enumerated in this sense by
- Let be the number of compositions with parts having as their largest part. Show that
where is the number of descents of .
- Deduce that
- Hence prove MacMahon’s identity
where .
Part 4: Proof of the Hook Formula
The restated Fundamental Lemma for reverse plane partitions
Fix a partition of size and let be the poset with elements the boxes of the Young diagram ordered by (the transitive closure of) and . Let be the sum of the largest parts of , and define a labelling so that the boxes in row of the Young diagram are labelled . (This labelling was seen for in the example at the end of the previous part.) Since this label is all-natural, the -partitions are precisely the reversed plane partitions of shape . By the restated Fundamental Lemma and the following remark we get
where the sum is over all permutation such that the linear order defined by refines . Equivalently, as seen in Part 3 and the final example, the label appears before both and before in the one-line form of (when the boxes exist). This says that and . Therefore, when we put in the box of with label , we get a standard tableau. Moreover, has a descent in position , if and only if appears to the right of in the one-line form of . Therefore defining , where the sum is over all appearing strictly below in , we have . We conclude that
where is the set of standard tableaux of shape .
For example we saw at the end of the previous part that the refinements of when correspond to the permutations
Their comajor indices are , , , , , their inverses are
and the corresponding standard tableaux, obtained by putting in the box of labelled are
Here the final tableau has strictly above and strictly above , so its comajor index is .
Hillman–Grassl algorithm
Since this post is already rather long, I will refer to Section 7.22 of Stanley Enumerative Combinatorics II or, for a more detailed account that gives the details of how to invert the algorithm, Section 4.2 of Sagan The symmetric group: representations, combinatorial algorithms and symmetric functions. For our purposes, all we need is the remarkable corollary that
where is the hook-length of the box . We saw the special cases for the partitions and above.
This formula can also be derived by letting in Stanley’s Hook Content Formula: see Theorem 7.21.2 in Stanley’s book, or Theorem 2.6 in this joint paper with Rowena Paget, where we give the representation theoretic context.
Proof of the Hook Formula
Combining the results of the two previous subsections we get
Equivalently, using the quantum integer notation and , we have
This is the -analogue of the Hook Formula; the weaker normal version is obtained by setting to get