Phil Lucht Math & Physics Archive
Home / Math and Physics Files / Math / Wedge World / Sjamaar Forms

proof of a lemma regarding Hodge star v 2 (CH2)

DOCX · 20.9 KB
Open DOCX file

Short note dated 7.15.15 by Phil proving the lemma *(*dxI) = (-1)^(kn+k) dxI, so that **α = (-1)^(kn+k) α for any k-form. It defines signs S and S' by calibrating against the volume form dV, proves a sublemma on commuting the complementary product past dxI by moving each factor through n-k terms, and shows SS' = (-1)^(kn+k). A closing comment derives an explicit expression for S'.

AI-written summary; may contain errors.

Extracted text (machine-read; may contain errors)
Proof of a Lemma PhL 7.15.15 Show that for a k-form: *(*dxI) = (-1)kn+k dxI Sjamaar p 28 Ex 2.12 (CH2) (*) [ If α = any k-form, this shows that **α = (-1)kn+k α .] 1. Definitions: dxI ≡ dxa dxb .... dxq a < b < ... <q a k-form (1.1) Πh dxi ≡ dx1 ....dxa ....dxb........dxq ....dxn red = holes (1.2) *dxI ≡ S Πh dxi (1.3) 2. Calibration for S: dxI(*dxI) = dV or dxI S Πh dxi = dV dxI(Πh dxi) = S dV (2.1) 3. Compute *(Πh dxi): (Πh dxi) = dx1 ....dxa ....dxb........dxq ....dxn *(Πh dxi) = S' dxa dxb .... dxq = S'dxI (3.1) 4. Calibration for S' : (Πh dxi)*(Πh dxi) = dV (Πh dxi)S'dxI = dV (Πh dxi)dxI = S' dV (4.1) 5. Sublemma: (this will be proved in a later item below) (Πh dxi)dxI = (-1)kn+k dxI(Πh dxi) . (5.1) 6. Therefore using (4.1) on the LHS of sublemma, and (2.1) for RHS of sublemma, we have S' dV = (-1)kn+kS dV or SS' = (-1)kn+k (6.1) 7. Now consider *(*dxI) = *( S Πh dxi) // using (1.3) for *dxI = S *(Πh dxi) // factor out constant = SS'dxI // using (3.1) for *(Πh dxi) = (-1)kn+k dx // using (6.1) (7.1) This concludes the proof since we have shown (*). 8. Proof of sublemma: (Πh dxi)dxI = (-1)kn+k dxI(Πh dxi) Start with the LHS (Πh dxi)dxI = dx1 ....dxa ....dxb........dxq ....dxn dxa dxb .... dxq Now let's start by moving dxa all the way to the left.. It has then to pass through n-k dxi factors, so we pick up (-1)n-k. We then have (Πh dxi)dxI = (-1)n-kdxa dx1 ....dxa ....dxb........dxq ....dxn dxb .... dxq Now move dxb in this same manner and pick up another factor (-1)n-k , so (Πh dxi)dxI = (-1)2n-2k dxa dxb dx1 ....dxa ....dxb........dxq ....dxn .... dxq Do this for all k factors up through dxq and end up with (Πh dxi)dxI = (-1)kn-kk dxa dxb .... dxq dx1 ....dxa ....dxb........dxq ....dxn = (-1)kn-kk dxI (Πh dxi) Finally, note that (-1)kk = (-1)k so (-1)kn-kk = (-1)kn-k = (-1)kn+k since (-1)2k = 1 Therefore (Πh dxi)dxI = (-1)kn+kdxI (Πh dxi) (8.1) and the sublemma is proved. Comments: Although the derivation involved signs S and S', we never had to obtain an expression for S or S' separately, we only needed an expression for the product SS'. In my original document I showed that S = (-1)a+b+..+q (-1)k(k+1)/2 note that S2 = 1 (1.13) for dxI = dxa dxb .... dxq a < b < ... < q a k-form (2) Since SS' = (-1)kn+k, we could then deduce an expression for S' S' = (-1)a+b+..+q (-1)k(k+1)/2 (-1)kn+k (-1)a+b+..+q (-1)-k(k+1)/2 (-1)kn+k (-1)a+b+..+q (-1)kn-k(k-1)/2