This blog post summarizes the 1996 paper by M. Zajac titled "A quasinilpotent operator with reflexive commutant," published in Studia Mathematica.
TL;DR
The paper provides a definitive answer to a long-standing question in operator theory by constructing a specific nonzero operator that is "quasinilpotent" (its spectrum is just zero) yet possesses a "reflexive commutant." This proves that even operators with the simplest possible spectra can have a rich, structured relationship with their invariant subspaces.
What Problem Does This Paper Address?
In the study of linear operators on Hilbert spaces, mathematicians are often interested in the relationship between an operator $T$ and the collection of subspaces it leaves "invariant" (subspaces $M$ such that $T(M) \subseteq M$).
A central concept here is reflexivity. An algebra of operators is "reflexive" if it contains every operator that leaves invariant all the subspaces invariant under the algebra itself. A particularly tough question, posed by Warren Wogen in the 1980s, asked: Does there exist a nonzero operator $T$ that is quasinilpotent (meaning its spectrum $\sigma(T) = \{0\}$) such that its commutant (the set of all operators that commute with $T$) is reflexive?
Previous attempts to solve this (notably by Herrero) were found to be incomplete. Zajac’s paper aims to provide a rigorous, constructive proof that such an operator exists.
Methods
The author constructs the operator $T$ as a weighted shift operator on an infinite-dimensional Hilbert space $H$, which is defined as the direct sum of infinitely many copies of a smaller Hilbert space $\mathcal{R}$.
The key technical innovations include:
- Dense Sequence Construction: The author proves a lemma showing that one can find a sequence of unit vectors $\{g_n\}$ that is dense in the unit sphere of $\mathcal{R}$, but where the distance between consecutive vectors $\|g_n - g_{n+1}\|$ vanishes at a controlled rate ($1/n$).
- Careful Weight Selection: The operator $T$ is defined by weights $T_n$ that act between the copies of $\mathcal{R}$. These weights are built using projections $P_n$ onto the vectors $g_n$.
- Logarithmic Scaling: To ensure the operator is quasinilpotent while remaining bounded, the author introduces a logarithmic factor ($1/\log n$) into the definition of the weights. This specific decay rate is the "secret sauce" that makes the proof work where previous attempts failed.
Datasets
Not discussed in the paper. As a theoretical mathematics paper, it relies on formal proofs and logical construction rather than empirical data.
Key Insights and Ideas
- The Power of Small Steps: Lemma 2 is a beautiful geometric insight. It shows that you can "traverse" the entire unit sphere of a Hilbert space using a sequence of points that are arbitrarily close to one another. This allows the author to "smoothly" transition the operator's action across the space.
- Commutant Minimization: By choosing the projections $P_n$ based on a dense sequence, the author ensures that any operator $A$ commuting with $T$ must essentially be a power series in $T$. This "shrinks" the commutant enough to make it reflexive.
- Spectrum Control: The quasinilpotent property ($\sigma(T) = \{0\}$) is achieved by making the "product" of the weights grow slower than any geometric progression, a feat accomplished by the $1/\log n$ term.
Key Results
The paper culminates in a theorem proving that the constructed operator $T$ satisfies three critical conditions:
- $\sigma(T) = \{0\}$: The operator is quasinilpotent.
- $\{T\}' = \text{Alg } T$: The commutant of $T$ is exactly the weakly closed algebra generated by $T$ and the identity.
- Reflexivity: The commutant $\{T\}'$ is a reflexive algebra.
This confirms that the property of being "hyperreflexive" (where the commutant is reflexive) is not incompatible with having a trivial spectrum, provided the operator is nonzero and acts on an infinite-dimensional space.
Limitations and Future Work
The paper is an existence proof; it constructs one such operator but does not provide a general classification of all quasinilpotent operators with reflexive commutants.
The author acknowledges that this work was prompted by a gap in a previous proof by Herrero. While this paper closes that specific gap, it opens the door for further research into the hyperreflexivity constant of such operators—a numerical measure of how "close" an operator is to being reflexive—which remains a topic of interest in functional analysis.
Summary updated