Reflection Positivity Then and Now

Arthur Jaffe

Dedicated to the memory of Robert Schrader

arXiv:1802.07880v1 [math.HO] 21 Feb 2018

Konrad Osterwalder and Robert Schrader discovered reflection positivity (RP)

in the summer of 1972 at Harvard University.1 At the time they were both my
postdoctoral fellows, and together we formed the core of a group at Harvard who
studied quantum field theory from a mathematical point of view.
RP has since blossomed into an active research area; this conference in Ober-
wolfach marks the 45th anniversary of its discovery. When Palle Jorgensen, Karl-
Hermann Neeb, Gestur Ólafsson, and I first envisioned this meeting in 2014, we
had hoped that both Robert and Konrad would be here. Unfortunately neither is:
Robert died in November 2015, and Konrad had to cancel because of a conflict.
So I have been called upon to make some remarks to set the stage.
These comments not only relate to the early discovery of RP, but they also
illustrate that RP is still an active and interesting area of research. The original
discovery of RP arose from an effort to relate two different mathematical subjects
in a specific way: what property is needed to start from a classical probability
theory of fields (i.e. a statistical mechanics of random fields), and end up with
a quantum theory of fields? Of course one wanted to have a framework that
would apply to the putative quantum fields that physicists believe describe the
interactions of elementary particles.
The original work focused into finding a set of axioms for Euclidean-covariant
Green’s functions, that are equivalent to the Wightman axioms for the vacuum
expectation values of relativistic quantum fields [42, 43]. The Euclidean Green’s
functions are objects in classical probability theory, while vacuum expectation
values describe the quantum mechanics of fields.
One can regard the relation between the analytically continued expectations
(in time) to probability theory, as generalizing the “Feynman-Kac formula.” This
formula gives a Wiener integral representation that provides solutions to the heat-
diffusion equation, namely the analytic continuation of the Schrödinger equation.
One side of the equation is purely classical; the other side is an analytic continu-
ation (justified by the positivity of the energy) of quantum theory.
An immediate application of RP was to provide a framework for the first exis-
tence proof for examples of non-linear quantum field theories with scattering [12].
These are quantum, non-linear, scalar wave equations with energy density λP(ϕ)2 .
Here λP is a polynomial bounded from below, and the subscript 2 denotes two-
dimensional space time. Theorem: For 0 < λ sufficiently small, these models
exist and satisfy all the Wightman axioms for a quantum field theory. The models
actually satisfy as well the more-detailed Haag-Ruelle axioms for scattering, which

1This is based on the opening talk on November 20, 2017 at the conference “Reflection
Positivity,” held at the Mathematical Research Institute, Oberwolfach, Germany.
1
include the existence of an lower and upper mass gap, leading to an isolated eigen-
value m in the mass spectrum. In addition they yield local algebras that satisfy
the Haag-Kastler axioms. For these reasons the discovery of RP marked a turning
point in the study of relativistic quantum physics.
Amazingly, the RP property not only arises in quantum physics. But RP con-
nects to areas of research in fields ranging from operator algebras, mathematical
analysis, probability, and representation theory on one hand, to statistical physics
and the study of phase transitions on the other. RP has enabled the proofs of nu-
merous interesting and deep results in far-flung areas of mathematics and physics.
Similar positivity conditions appear in other subjects, so one might dream that
new areas of relevance for reflection positivity will appear in the future, in other
mathematical areas.
Since RP is now an enormous field, I apologize in advance for my sparse list of
references; these are only meant to be personal and impressionistic, and they cover
a small selection of papers from an enormous universe of possibilities. However, I
am confident that the other speakers during this week will fill the gap by citing
many other important papers.

1. The Background Story

The analytic continuation of the expectations of fields to imaginary time was
proved in the general framework of Wightman in 1956 [53]. This result is a con-
sequence of the assumption that the spectrum of the energy H is non-negative,
and that the time-translation automorphism arises from the unitary group eitH
acting on the Hilbert space H of quantum theory. Furthermore one assumes that
the Poincaré group acting on H has an invariant vector (a vacuum).
This analytic continuation is the non-perturbative parallel to “Wick rotation”
of time introduced in the physics literature to make sense of the terms in the
perturbation expansion of a Lorentz-invariant theory [51]. Jost studied properties
of analytically continued Wightman functions to Schwinger functions, including
their symmetry at imaginary time, as detailed in his 1957 book [30]. Schwinger
emphasized the notion of Euclidean quantum field theory in his well-known 1958
paper [46]. According to Miller [37], Schwinger claimed to have discovered these
results about Euclidean fields seven years before he published them.
In any case, the advent of Euclidean expectation values and Euclidean fields set
the stage for the pioneering work of Kurt Symanzik, who proposed in 1964 that
the Euclidean field could be formulated as the classical Markov field. Symanzik
described these ideas in preliminary form in his widely-circulated Courant Institute
Report [47] and in the paper [48]. He also presented a monumental course in his
1968 Varenna summer school lectures [49].
The Varenna lectures mark the end of Symanzik’s research on Euclidian quan-
tum field theory, which he explains in his introductory lecture. There he estimates
the efficacy of the different approaches known at the time for constructing an ex-
ample of an interacting field. Unfortunately Symanzik under-estimated the value
2
of his own approach. In fact when it was understood some five years later, it
turned out to play an important role.
Symanzik’s approach cried out for building a mathematical foundation by which
to understand it. His vision appeared to be a promising blueprint for finding a non-
trivial field theory. Symanzik made a valiant attempt to implement his program
for the ϕ4 theory in two space-time dimensions, by working on this problem with
S.R.S. Varadhan. But the gap between their results (described in the appendix
to [49]), and the much more extensive results necessary to establish the existence
of an interacting Euclidean field theory—even in two dimensions—remained im-
possible to bridge at the time.
Furthermore, even if one could solve that problem of constructing the probabil-
ity distribution for a Euclidean classical field, there remained another fundamental
question: how can one relate a classical Euclidean field to a relativistic quantum
field? In other words, as there is no Hamiltonian for the Euclidean classical field,
how can one justify analytically continuing back from Euclidean time to real time?
Namely, how can one justify an inverse Wick rotation? In the rest of this work I
will call this question the reconstruction problem; similarly I call a solution to this
problem a reconstruction theorem.
How to prove a reconstruction theorem baffled mathematical physicists during
my student days, and it became the center of much discussion. But there were
no viable suggestions of how to resolve it. The same question also baffled par-
ticle physicists, including Schwinger2 who regarded it as important. This latter
statement is documented in Schwinger’s response to a question asked by Pauli
about understanding spin and statistics in the Euclidean framework; it followed
Schwinger’s talk at a 1958 CERN conference. “The question of to what extent
you can go backwards, remains unanswered, i.e. if one begins with an arbitrary
Euclidean theory and one asks: when do you get a sensible Lorentz theory? This
I do not know. The development has been in one direction only; the possibility of
future progress comes from the examination of the reverse direction, and that is
completely open.”
The first big step in the reconstruction from Euclidean to real-time was achieved
in 1972 by Edward Nelson. He had been fascinated by Symanzik’s Markov field
approach, and formulated that framework in mathematical terms. Nelson discov-
ered a reconstruction theorem that starts from a Markov field and yields a scalar
bosonic quantum field [38]. He then showed that the free (Gaussian) Euclidean
field satisfies the hypotheses of that construction [39].

2In those days, discussions after lectures at important conferences were recorded or tran-
scribed; then they were published along with the lectures in the conference proceedings. In
perspective, the discussions are often of greater interest than the talks. I was grateful to learn
in [37] about the reference [45], where one can read this fascinating exchange. The book that I
quote can be found online by following the hyperlink to CERN, and the quoted passage appears
at the end of page 140.
3
 2. The Discovery of Reflection Positivity
This was the situation in the summer of 1972 when I began to read the preprint
of Nelson’s reconstruction paper. This led to several discussions with Konrad
Osterwalder on the topic. Not only did I wish to understand Nelson’s construction,
but I hoped that one could discover a more robust reconstruction method that also
applied to fermions. In the Euclidean world fermions are Grassmann, rather than
abelian, so they did not appear to fit into a Markov setting.
Furthermore, the Markov property that Nelson assumed was a strong one, and it
had not been proved for the examples of two-dimensional interacting fields known
at the time, although Nelson could verify it in the case of the free field. Konrad
was looking for an interesting project to work on, and I thought a new way to do
reconstruction could be very fruitful. I was deeply engrossed at the time in trying
to understand aspects of the three-dimensional ϕ4 theory.
Meanwhile Robert Schrader was on vacation, visiting his friend (later wife) in
Germany. Since one could not park a car on the street for more than one night
near the Cambridge address where Robert lived, I had suggested that he leave his
car in Northampton, Massachusetts. This is a small college town where my wife
worked, and I knew that there were no overnight-parking laws like in Cambridge.
Just after Robert returned, I planned to drive to Northampton, and from near
there to fly to Chicago for a conference. So I took Robert with me to pick up
his car, and for nearly two hours we discussed the reconstruction project while I
drove.
I explained to Robert what Konrad and I had been thinking about, and I
encouraged Robert to think more about the question. But Robert immediately
resisted. He too had seen Nelson’s paper and thought that Nelson’s construction
was very natural, leaving little new to be discovered. As the trip progressed, I was
getting more and more upset—and afterward I recalled worrying about paying too
little attention to the road. For not only did I feel that the question was extremely
interesting, but I really hoped that Robert and Konrad could make some progress
on the problem while I was in Chicago. Finally I seemed to break through, only
shortly before we arrived at Robert’s car.
About one week later, when I returned to Cambridge, Massachusetts, I found
Konrad and Robert enthusiastic, proud, and delighted. They thought that they
knew the key to a new reconstruction method: they had discovered the reflection
positivity property! A reflection-invariant functional ω on an algebra A = A− ⊗A+
is reflection positive on A+ ∋ A with respect to the antilinear reflection homomor-
phism Θ mapping A+ 7→ A− , if 0 6 ω(Θ(A)A). And in their example, the RP
form provided the inner product space a Hilbert space H for quantum theory.
Their original celebrated publication [42] appeared a few months later. This is the
statement for bosons; more generally one uses a twisted product Θ(A) ◦ A that
reduces to Θ(A)A for bosons, see [42, 18, 20].
The RP form actually provides the quantization of the classical system. If ω
is invariant under a time-translation *-automorphism σt for t ∈ R, and αt acts
on A+ for t > 0, then the quantization of αt yields a positive Hamiltonian H on
4
H. The importance of RP in quantum physics stems from the fact that the inner
product and the Hilbert space for almost every quantum mechanics or relativistic
field theory arises from the quantization of some classical system that has the
reflection-positivity property. Often one calls this property “Osterwalder-Schrader
positivity.”
It was discovered that the original Osterwalder-Schrader paper contained an
error involving continuity for a tensor product of distributions. Although some
persons claimed that the gap was serious, I was sure that the problem was minor;
that is what turned out to be the case. They corrected their method in a second
paper [43], where they also gave a stronger version of their reconstruction theorem–
showing the equivalence of their assumptions to a modified version of Wightman’s
theory.
It is also meaningful that Edward Nelson explained in his 1973 Erice Lec-
tures [40], how his Markov field fits into the Osterwalder-Schrader (OS) frame-
work, even though at that point the correction to the OS method had not yet
been found [41]. But by that time, Nelson had come to regard the OS reconstruc-
tion as the natural way to pass from Euclidean theory to quantum theory.

3. The Proliferation of Results Related to Reflection Positivity

Since the appearance of RP in quantum field theory, it has permeated many
other fields. Here we only give an impressionistic view. Robert Schrader left his
position as postdoctoral fellow at Harvard in 1973 to go to Germany. His successor
as postdoctoral fellow was Jürg Fröhlich, who became one of the first persons to
investigate RP in great detail. Right in the beginning he made an interesting study
of the generating functionals and the reconstruction theorem via RP-functional
integrals, both the zero-temperature and also finite-temperature fields [6]. I refer
the reader to the notes of his presentation at this meeting for several of the many
other perspectives that Jürg pursued [7].
It also turns out that RP has close ties with preceding work: including with
Widder’s investigation of the Laplace transform in the 1930’s [52], as was discov-
ered by Klein and Landau [32]. Furthermore, the reflection Θ in RP is related in
various ways to the reflection J in the Tomita-Takesaki theory of operator alge-
bras [50], as well as to other reflections in its predecessors. Some references on
these latter connections can be found in the introduction to [19] and of course the
talks of Fröhlich [7] and of Longo [36].
3.1. Phase Transitions and Symmetry Breaking. Physicists believed for
many years in symmetry breaking and vacuum degeneracy for certain quantum
field theories. The idea of the proof goes back to Peierls’ analysis of phase transi-
tions in the Ising model in 1936, but it was completed mathematically by Griffiths
and by Dobrushin some thirty years later.
One expects that a λ2 (ϕ2 − 1/λ2 )2 field theory with a “W-shaped” potential
will also have a phase transition for the real parameter λ sufficently small. In
this case the field will behave much like an Ising system at low temperature: the
average of the field will be approximately localized near the two minima of the
5
potential at ϕ = ±λ that are separated by a large potential barrier of height λ−2
at ϕ = 0. This became a much-sought-after result, which was announced without
proof in 1973 by Dobrushin and Minlos in a widely-cited paper [3].
Another early application of RP that one might not have anticipated, was the
important use of the Schwarz inequality arising from RP to establish global es-
timates of local perturbations—by reflecting the perturbation multiple times. In
particular this was an important component in the first mathematical proof of the
existence of symmetry breaking and vacuum degeneracy in quantum field theory.
One used multiple-reflection RP bounds in λϕ4 theory to estimate the deviation
between local fluctuations in that model from a Peierls’ type estimate, yielding a
degenerate ground state for λ sufficiently large [13].
3.2. Statistical Physics. RP played a major role in lattice statistical physics.
There is a very large literature, with early work by Fröhlich, Simon, and Spencer [10]
on continuous symmetry breaking, by Fröhlich, Israel, Lieb, and Simon [8] on es-
tablishing RP, and by Dyson, Lieb, and Simon [4] on establishing phase transitions
in quantum spin systems. Lieb used these methods to analyze the ground state
vortices in some models [34], see also [2]. One can consult the review of Biskup [1]
for much other work.
3.3. Relations to Mathematics.
3.3.1. Representation Theory. The first work relating RP to representation theory
for quantum fields arose from the desire to obtain representations of the Poincaré
group from the analytic continuation of quantization of the representations of the
Euclidean group. After the initial work of Osterwalder and Schrader [42, 43], this
was investigated abstractly by Fröhlich, Osterwalder, and Seiler [9] and by Klein
and Landau [32, 33].
This was eventually developed into an entire subfield of representation theory
and stochastic analysis studied by Klein and developed extensively by Palle Jor-
gensen, Karl-Hermann Neeb, and Gestur Ólafsson, as well as their collaborators.
See [31, 28, 29] and the citations in these papers.
3.3.2. Relations to PDE. For a free scalar field, the RP property is equivalent to
a statement about monotonicity of the Green’s functions with respect to a change
from Dirichlet to Neumann boundary conditions. Let C = (−∆ + 1)−1 denote the
Green’s operator for the Laplacian on Rd , and let CD denote the Green operator
obtained by imposing vanishing Dirichlet boundary conditions on the time-zero
hyperplane. Let CN denote the corresponding Neumann Green’s function for
vanishing normal derivatives on the time-zero hyperplane. Then RP is equivalent
to the following statement of operator monotonicity on L2 (Rd ),
CD 6 CN .
This can be seen by expressing CD and CN using the method of “image charges,”
familiar in physics [11]. One can generalize this to obtain a condition for RP on
reflection-invariant spaces Σ = Σ− ∪ Σ0 ∪ Σ+ with an involution Θ that exchanges
Σ± and leaves Σ0 fixed.
6
3.3.3. Fourier Analysis and the Inequalities. For the approach to classical Fourier
analysis through RP, see the contribution of Frank to these proceedings [5]. It is
interesting that it is possible to prove Fourier bounds in subfactor theory to prove
uncertainty principles, and this is related to the picture analysis that I describe in
§4 below [25, 26]. This area of research on the analytical inequalities for pictures
in a non-commutative algebra is just emerging; it seems to offer potential exciting
new insights.

3.3.4. Relations to Tomita-Takesaki Theory and the KMS Property. This is an-
other enormous subject. The reflection Θ arises in several different ways as the
Tomita-Takesaki reflection operator J. It is also related to the KMS property,
discovered by Haag, Hugenholtz, and Winnink [14].

3.3.5. Stochastic Quantization. The study of classical stochastic PDE’s has re-
ceived a great deal of attention recently. Physicists have proposed that a classical
equation with a white noise linear forcing term can be used to quantize a classical
equation—the method of stochastic quantization. The quantum distribution arises
as the limit of the distribution of solutions at infinite stochastic time. Unfortu-
nately reflection positivity does not hold for the distribution of the stochastic field
at any finite stochastic time [15].3 Therefore it is difficult to see how to use this
method to obtain a limit that satisfies RP–except when one can analyze the limit
in closed form.

4. Some Elementary Remarks on the Mathematics of Pictures

When Zhengwei Liu came to Harvard as a postdoctoral fellow in the summer
of 2015, we spent a good deal of time telling each other about each of our current
areas of research. After a while we realized that we could combine our work,
by defining a pictorial framework that we called “planar para algebras.” This
led us to a new way to think about the RP property. We implemented all this
in the framework of statistical mechanics models of parafermions on a lattice—a
setting we call PAPPA. In these examples, we found a geometric interpretation
and proof of RP [20] for a very wide class of Hamiltonians. We then found that our
models could be viewed as languages; we had first used this principle in joint work
with Alex Wozniakowski on quantum information [22, 20], and later found that it
provides very interesting insights into other fields of mathematics [21]. Hopefully
they will also contribute in the future to physics.
Parenthetically in the case of parafermions of degree d = 2, the fermionic case,
elementary parafermions are called Majoranas. As quadratic functions of Majo-
ranas represent classical and quantum lattice spins, the RP for parafermions leads
to RP proofs in these cases [24, 18, 19].

3We have only proved this for a linear field. One sees in this case that RP does hold in the
infinite-stochastic-time limit. If the Wightman functions for a non-linear theory are continuous
(as expected) in the perturbation parameter, then RP will also not hold at finite stochastic time.
In that case, however, one needs an independent method to establish RP for the limit.
7
 I believe that the simple fundamental idea embodied in our new pictorial ap-
proach to RP will prove fruitful in many other contexts. Let me explain some
central ideas that lead to RP, without much elaboration. Planar algebra is based
on the mathematics of a ∗-algebra of pictures. Multiplication is vertical compo-
sition of pictures, with algebraic right-to-left order represented by top-to-bottom
composition of pictures. For a picture T , one represents the adjoint map T 7→ T ∗
in the algebra by vertical reflection of the picture. If

R
R= R , then R∗ = .

Composed with multiplication, the vertical reflection gives an anti-linear anti-

homomorphism that reverses the order of multiplication of pictures,
 ∗
 
  R
K 
 K 

 
RK = , ∗
(RK) = 


 =


= K ∗ R∗ .
  K
R 
 R 

 

It is natural to consider a second reflection Θ that is horizontal. This reflection

defines an anti-linear homomorphism of pictures,
 
 
 

 K 
 K
 
Θ(R) = R , Θ(RK) = Θ  = = Θ(R)Θ(K) .
 
 
 

 R 
 R
 

The basic idea of the geometric notion of RP arises from the fact that the two
Θ-reflection of pictures is related to the ∗ -reflection of pictures by a 180-degree
rotation that we donote Rotπ . Algebraically Rotπ (Θ(R)) = R∗ , while in pictures,
 

R
 
Rotπ  R = .
 
 

The horizontal reflection is the reflection for RP. The vertical reflection, on the
other hand is associated with the fact that R∗ R is positive, if the pictures have
a positive expectation defining a non-negative form that can be used to define an
inner product and a Hilbert space through the GNS representation.
In order to understand the geometric proof of RP, one needs to analyze the
horizontal multiplication of pictures. This horizontal multiplication defines the
8
convolution product of operators,

T ∗S = T S .

As there are two types of multiplication, rotation of pictures by 90 degrees is a very

important transformation. Denote this by Fs ; in terms of the input and output
strings, this permutes them cyclically by one:
.

Fs T = Fs T = T

We named Fs the string Fourier transform (SFT) as it generalizes the normal

Fourier series. The SFT sends a product of neutral 1-qudit transformations to the
convolution of two SFT’s,

Fs = Fs T Fs S

T
.

For a transformation with two input and two output strings this is a rotation by
90 degrees.
One needs a state on the algebra of pictures, in order to apply the GNS con-
struction and to recover the identity of pictures with elements of a Hilbert space.
This arises from attaching the input strings to the output strings, and it defines a
trace functional on the algebra of pictures.

4.1. The Geometric Interpretation of Reflection Positivity. One visualizes

the new geometric proof of RP from looking at Figure 1. This illustrates how one
relates the horizontal reflection Θ(R) of R with its vertical reflection R∗ . The
circle product has the pictorial meaning that Θ(R) and R appear at the same
vertical level.

tr(e−βH Θ(R) ◦ R) =
bSHKdgDCvZgICqtIqcr26YKrw
SvtYQr2gII9GIhKq8jpyrapwq
JW4EkZ5Ulw9vsclwrnFlLlCs6aIIH
cG
0aSuQ5KyyZHibXY5rhTNriXJFB03w
WB
y8II3noD4r41+R2KChIPGtV02FQzIkqNGF
VDdYpA
+IFYSQP/UER/5rcDgUFiWetajoMijlRhSb
iOoGi/
TjYsvIwsbg72oU9six/ARY7ARgxZvyhUtF9Dkn
TyYgh0sajgb
hqnHxZWRhY7AXdWpb5Bg+Ygw2YtDiTbmi5QKan
BWXgyBLhYV3
aJ6Fkpy/Smf69OJGesJ/gfJ/g7i/VzaM4kHtG1ANnk
JBkm5xoWt3A2NAU
V5Uz0JJrj/lc306MWM9wf8g2d9BvJ9LeybxgLYNyCa
hCTD5FzD4hbOhqZ
HlNqgZ2aaDjS0uiSk62wvZMkYyca6thlliam6eifVOT0HQ6
CwSlCD4ZnDDyQsC/OI85
9kFIb9MxM04GGVpeEdJ3thSwZI9lY1zajLDFVV++kOqfnYI
MUVglqMDxz+IGEZWEecX
WOWdTmPvdK7rnfA2BKjiX23h
BKu802nsnc51vRPehgBV/Kst
VXXYBXDkAPfXvi+ruVEGJ8TxcrlA
H
vgqgvwygHooW9PXH+3EkKM7+lihb
q
WydfrKZeFJv3kD9iDX0Z6AuSM52cWGhuH
8hCFiG
bevk89XUi2LzHvJHrKEvA31BcqaTEwvNz
fhzBED
PfRUIF5CAR6ZYZHfLnDibDvtwV8WbzJP5A6wm
4y6CxQ+XnS
7qu0iogBwkIt0yo0P+3MFk2Je7It5snsQfaD1
5ZRd4HCx4u
nax8EC0Atg+YeoHrRP4VOrCdwEOyAufMDmwnQlbkkN
dU28eL5FigYdoA
o5OVD6IFwPYBUy9wnci/Qge2E3hIVuCc2YHtRMiKHF
3e+rN4yVSLPAQD
7ku4fLwLVnE999OTl/jPtffukGobeYH0RXS3c8s0/n3onn
ez43cN/WdOFs5q588jx
/y3cNl4Nqzie++nJw/xv0vv3CD0FvMD6KrpTue2adz78Rz
p4dv3v4b2u6cFYzdx45
b4fKp7/0Nfe1TUaBPBVqgP6
<latexit
b4fKp733qn+hTUaBPBVqgPy
sha1_base64="tI8dItQBIMVYQ6mqgAimKmCpUSA=
sha1_base64="72RT8hSQEVLatdtYvPbI+7ptZk8= oKSKRW2IkZX2Ut/p6JQS0oXlXxzedCV+Z6aiT88q5eBXXZEluakLIlJ6typM7RZkW08mWV5hhC9aA+o5VkirkVy+rXIckoOac0rE
uCAhKpFXZixldZi79nYlALildVfPO50JW5nhoJv7yrV0FdtsSWJqRsycmqHKlztFkRrXxZpTmGUD2oz2glmWJuxbL6dUgySs4pDS
67muI2Kueq2cwl6gVtgcuV7ikmVhFvelKxqeH9fqLMSppuAvtgoLAxIrSjmaYDAbBTVSLlKRmd9nMfh5c0BmTTH3t8HnNGUuVg
SetJluF0+mZ7RXbPoRxI9xupPuwQnp+48vDZEbPQzKkabKCpLMf0GeI+S+o1tpnP5GVLMUg9KGMj0U0jCqpMaghqGrya2rA2TESK
YD3XdUTMVa+VU9gL1AqbI9dLXLIszOK+dEXD8+NanYU41RT8xVZhYUBiRSlHEwxmo6BGykUqMvP7LAY/bw7IrCnm/jb4HU2Zi9
QdJ6kmU4nb7ZXpHdcygH0v1Gqg87hOcnrjx8dsQsNHOqBhtoKssxfYa4z5J6jW3mM3nZUgxSD8rYSHTTiIIqkxqCmgavpjasDROR
2iJAyEak3qkC5GVMvJTd3RFyOkptF9cDcfKoEzmcWT8XHcDHHmpmpTCk1E1zPTIZUm5KYg0bKqB2v9jw/i3lQbqY2nK5F
cMqwo5SsIYa0tUiFVatbiLJgC/xrezxGYuC6Swnb3hWgJmF3/vOH6Y0Mt/XrwCcSFqmldkA/l8VBQCD9AI9z2a4oiJBpeaaiRx6U
MDhFSJiL1RhUoN2PqpeTmjojLUXKzqB6Ym0+VwPnM4qn4GC7mWDMzlSmlZoLrmcmQalMSc9BIGbXj1Z7nZzEPys3UhtO1
SAYVUhR0kYY22JCrFKqxZ30QTg1/h2ltjMZYEUtrM3XEvA7OLvHccPE3r5z4tXIC5ETfOKbCCfj4pC4AEa4b5HUxwx0eBSU40kLh
EcMnvbSOMoI6ndBNz/p7SRupzM+5wJC11nMp3kDa9EOFb8D9T08qUguUbQDbL68j/dFZ6gYOSxGGvxwgbXe5eBzAL
lCiPk+UBwa+duB+sxLEpEQ9FQgJVk4JkEtJ+vQaBfUqBWETBQ1RNpb9oEI0qQMaeiBr5+CZ1F7gly+nHCtpuOlMxBTTGF+FTdEKB
POIwdNeGkcZQf0u6OYnvZ3E7XTG51xA6DqL+TRvYC3aoeJ3oL6HJxXJJYp2gM2X9/G+6AwVI4fFSIMfLrDWuxx8Do
A1K1OeJ8sDA1w7cb1aCmJSopwIhwcopAXIpSZ9eo6BepSBsoqAhysayH1SIJnVAQw9k7Rw8k9oL/PLllGMlDTediZhiGuOrsClag
sAeGQrZLQYMk0SqeTk2IpHUhooECjk9YhR0JxvOOHbHI2YFhwWNrsGGbSFflrD1GgkEEUxdKMCNkrF9GhOyQk
icq4NeaJeMZJsOSB3DrKGYU7Qk6H0eRRDRDFElD6LwqlPVEH7Dqp/ad8hijcorMXvAkh7P0WR7AJECXUWSGQ4FBkST8s4fljDcSh
PJSAcslUSGiyZRul0clIspQMJDRRodNI65Egojnf8kE3OBgwLDkubHcNMuiJ/7SEKFDKIoliaESF75SI6dIeE
ARM61IU/UK0aSLQfkzkHWMMwJejKUPo9iiCiGiNJnUTj1iSpo30H1L+07RPEGhbX4XQBp76cokl2AKKHOAokMhyJD4mkZxw9rOA5
wUpxDsDAXKoUC5zdDquC/rouRwr48bB/irhG3Qzn7eSeOz06PtpnSQSTt74qABcUzaRqcBbdPlsubtnK
v4YyL3hyY6RFN7RBNdolX5RJM6Rav0iqZyi1bpF63kGJyOCH7BTSq3aFK/aJWO0UTPpFmYyJPzS+UU/BVOzyFeMFdzsv13pyBCG
iV4hSCnaFAORQotxlaHfdlXZQc7vVx4wB/lbAN2tnPO2l8dnq03ZQOMmlnTxw0II5J2+g0oG26XNa8nR
4Tf0zk/tBEh2hqj2iiS7Qqn2hSp2iVXtFUbtEq/aKVHIPTEcEvuEnlFk3qF63SMZromTQLE3lyfqmcgr/C6TnEC+ZqTrb/7hREa
SWB70xAZF62eV+j/pYG+jessBale6UJAr3TyZSDrW2cR6FQuAoJARKk0/FnjYdshu5i7nFFN3G58gK9PHfgBYUjOTHrGXB2pUTPq
PAlsbxoi46LV8wr9vzTQt3GdpSDVK10I6JVOvgxkfessAp3KRUAQiCiVhj9rPGw7ZBdzlzOqiduND/D1qQM/IAzJmUnPmKsjNWpG
WksMlogKeSdZKe4wUnxgGNnN5DR2YSJ1yvZtAORJIRbPUTxi6IBI/tomVe2zI7mBzZh6x98d2twd0kSacdL1UuT38c82gIi0AxK
2CkhqaLPMkxsgRoQQ2yk529I/gUP8I8ycTHkNRNDrzeQHC25FAW0MEOxfBVIjQu2cwQpiVQvR2yXGHHcoLdc9gxLpXkJFx5DAFYq
KAFDYZDfBUsk7Sc7zgxDigkdN76MhM4oTr1QzagUgygq1+wtgFkeCxXbTMa1tmB5Mj+5C1L767NbibJOm046XK5emPYx4NYREoB
VPUFJDk2WexBg5IpTARtnJjv4RHOofYf5kwmMoikZnPi9AeDsSaGuIYOcimAoRevcMRgizUojeLjnusEN5oe4ZjFj3CjIyjhymQA
j9K2tttwg3sDFBnVxvey9jRc2jgho9FMlMgG1CqlMYUaEwOKiUyYDES4KWKxDTY6RdyILYpcY+h1bQdyJ2C9DyTPaUWIPo
AIILwcy+WLqAjtKNQbafUIY3z24mzDmgQNRNMZgHzMocEja+ymK5UwXUYihMvPoonw6CyQyHIoMoa7g+GENx6HIMUPaRp82nsDk
eNHSXu7TbiBnQHq7GrDexk7eg4N3PCxSGYKZANKldKYAo3JQaVEBixGArxUkZgGO/1CDsQ2Je4xtJq2A7lTkJ5nsqfUAkS
EMA4eVAJl9MXWBHqcZAu08I47sHdxPGPHAgisYY7GMGBQ5Jez9FsZzpIgoxVGYeXZRPZ4FEhkORIdQVHD+s4TgUOWZI2+jTxhOY
j57hVbAy9r1yMGVs4ePFZjGMf6ZpN1fDcWyYWOkaxhmmQjqJ+rxA5Dt/JJH6VYGFw/ITRle+Cl0/3RxgFXHjT6Gy/t
FI3DMkvYF2n4FLRRooUjLCzgN7IuXT5NRb9yOkaORq3HL6OLMi9zuy6djAHa2jF2gaei7t/sxQ7KFzM33JWXoAyX2QQ04YS7i4w/
h9By3io2h96WLMWMLBy8+i3HsI12zqRqee8vEQsco1jAN0lHU7xUix+E7mcSvEiwMjp8wuvJd8PLp/gijgAtvGp3t9
ucASOWdK+QNuvoIUCLRRpeQGngX3x8mky6o3bMXI0cjVuGV2ceZHbffl0DMDOlrELNA1993Y/Zki2kLn5vqQMfaDEPqgBJ8xFfPz
q+exLtI829eRcE3ulZejKOT93FzI2CK07HOPLOv+K1tn1/HF8KZKeBvTzznEvUOncvnMVsZs+nsTWZJSNzjL7
NiC728ZloWPeu9vmejbctGAnELE+fJECWRZAEMhIPynAVQyMlRI+MuoQ3CXCf4Tev5iLnTZ88ij8PEW7CKxEoEAHxCCEekfJ22+3
Lbd0+ifaS5N++CwDs9S0/G8am7mLlRcMXpGEfe+Ze81rbvj+NLgew0sJdnnnOJWufuhbOYzez5NLYms2RkjtG
QBb8P0tw7LwUe/9LRN9Wy4akFOIOF+cKIEsC2AoBIT/NIBKRoYKCX8ZdQjuMsF/Qs9fzIUuex55FD7egl0kViIQ4ANCMCL946TN9
2BFqOr6zY0XuZRRHQbLj/j7WrIkb2eBJYh2coYOd57uW4wUOeL67f3xnS9/TyQeUD2B6sPXgPavzp1ar9
7TAcQX0ZjWM7iDzHd5O2tQrdpe2c26fuCB3O7ZkbjmOiUgK2UcsUnCwC9H8eAdLKjojtWRhezSaIcmZHZ6HYhxtlfaNVdDIcx958
7Ag1Hd/ZsSL3MoqjINlx/xBr1sSNbPAksQ7O0MHO813L8QIHPN/dP76zpe/p5APKBzA92HrwntX581cPr
vh2GI6gvo3FsB5Hn+G7Stlahu7Sdc/vUHaHDuT1zw3FMVErANmqZgpNFgP7PI0Ba2RGxPQvDq9kEUc7s6CwU+3CjrG+0ik6G49ib
">AABaAHic7Vxbb+PGFVbSW6reNu1jXwbxOrBXosOhKK2SwMBeEHRbbIPNxtu1V1QNiqJtwtQlJBX
P/pnPhZWY4Q1JeO0lbCbsWOXPmO9cZnjkUOVn6Xhjp+r/fevs73/3e93/wzg/bP/rxT376szvv/vx34WIVOO4LZ+EvgsOJHbq+N3
">AABaAHic7Vxbb+PGFVbSW6reNu1jXwbxOrBXosOhKK2SwMBeEHRbbIPN1tu1V1QNiqJtwtQlJBX
P/pnPhZWY4Q1JeO0lbCbsWOXPmO9cZnjkUOVn6Xhjp+ldvvf2d737v+z9454ftH/34Jz/92Z13f/77cLEKHPeFs/AXweHEDl3fm7
wY
sT
ksoKDPlwIJFObgM6ci1LmGS1r
oqSyAkM+HEikk9uAjlzLEiZZb
A+arIlDhksJdAMFQWoIaF
3IR022xCGDpQSaoaIANSQ
FkxiO0rRjIzK4/9E
SArBrF9xUhmZuWxf
yzJDSIM2CJg1
FjhlSGqRZ0KR
6blIohI8
MO0XAQRK
Hu2
3aP
iMTPa1o/aSxC06pPHbIHJ565Al
yVkcm+dtRekrhFhzR+G0Qubx2y
SPFpb4OZa4erwMWXS7Bz2Q
HCk+7W0wc+1wFbj4cgl2Lr
H44RMtBMcQu2YWCFu
heJyQiXaCQ6gdEyvE
w/sZATHAB3Z81
/uBnYzgGKAje3
uYrcuUMZ
L1eRO3co
dfRF
svIi Sj0368d+Rfhzf2YLiK47LBy+MvQ/39M/0rQePWvTzTuu
b9GOTfvx3pB/Hd7ag+Irj8sELY+/DPf0zfevBoxb9vNP
6UHaCSJMklYElZPF45dHwJNjJ2zVplFIkJ6e+TxIsC3JT38o
ZOXbQRK/fu2A18B5/TppA
kfpARpJokwSloTV04Vj14dAE2PnrFVmkYiQ3h55vAjwbUkPv
PVn5dpDEr1874DVwXr9O2
L3Hn5GXmRSF7ElU0sNkjGq3F2g
v0vYefkReZ1EVsydRSg2SMKncX
WXaXKkhKQ6uYQgnXBosJHydP0MY5fc
EB0
Z5bdpQqS0tAqplDCtcFiwsfJE7RxTl
l8U
APJdHHM/KHNGBCf6fAi03eVqGSm5NpKJGa
BS5uYih
XoQ0n08Yz8IQ2Y0N8p8GKTt1Wo5OZkGkqk
ixFbi5i
SWxThXfrqRRvrepkQe+K6xhJhAyDDVTMWqzGlBY
zJV/b8yWN4qG
dBLbVOHdeirFW6s6WdC74jpGEiHDYAMVsxarMaW
zfMnX9nxJo3j
AipGGFZDdSfakvvvTKxZd0LQ3ccOVHlRCSCoxTqsAUq
7L7snXEDTIqQf2mw
u4CGlYIdmNVF/qiy+9cvElXUsDN1z5USWEpALjlCowR
YnsvmwdcYOMSlC/q
2NU3U5eYNK5LjAjDfcRy/AY1ihRl4q/cHOb6BQuJTNcrXsiU
KomLytVdUlM66kfKKOsq
Bnb1zdQlJo3rEiPCcB+xHL9BjSJFmfgrN4e5fsFCItP1ihcy
toqiYfO0VFaWzbqS8oo6 XrfdaOy3Yut960HrSetZ60XJaf2z9rfWP1j+P/nD056O/HP2Vkr79VjrmFy3uc/T3/wDzOpfL</latexit>
6Zeu91k4Ltu63HrSetJ61XrSc1p9af2/9s/Wvoz8e/eXor0d/o6Rvv5WO+UWL+xz94z/6mJlR</latexit>
sV6Scr2TBkr3a1Snr+T2tjqB2o3PvTRTnE9dZqsk7IglVpCsZ1tbx/YE9+VdHDSP0arIyEkctNXZ0b2
+sMv4qxDbI1BDH/BC17O+EukgyMdKTl5Tgjy4gPERV++WZ6fsSfMtd91GGAoudTjaypuPVTAg8xeNpJXuGZyZi+zklqgN+s5ud1+
2ixXpKyPVPGSnerlOfvpDa2+oHajQ99tFNcT50m66QsSKWWUGxn29sH9sR3JR2c9I/R6kgIidz01ZmR
sA6w69ibINsDUHMP0HL3k64iyQDIx1peTnOyDLiQ0SFX76Znh/xp8x1H3UYoOj5VCNrKm79lMBDDJ52kld4ZjKmr3OSGuA3q/l5n
RPUxZSTPUHMyXt7DtWeXde5GKIcu/P+EHFaGDGO7Ylyz+JCxpXCRHZwKnOXyVMfNBfOX7RdYk7eq4TAqqZjJITsrRVFhg5rwmTQ
KFF9TBnJM9ScjJf3cO3ZZZ27EcqhC/8/IYeVIcPYrhjXLD5kbClcZAenAme5PNVxc8H8ZfsF1uStajiMSipmcsjOSlFU2KAmfCb
3Ys2ARLdBGOsEPw3+GX/a28iLw4PE+SmvdGXjwfN9AXwB1+mHkOecx2gjPkwS8D6wpbbC+xJ5MgPXFBW5t/ne28iPv1I7cuJf
0gLYsVw6BirMv29GwRRim/Snx4G/jowIkCH4ErRmN7vEryw8MSHYs/cyM3wAfOYh4FCx8dXOD/lyNjnHHLekSjwyR+xIAroaRtvK
8z9ixYRAu0kU7ww/Cf4Ze9rbwIPHi8j9JadwYePN830BdAnX4Yec55jDbC8yQB7wNrShusL7AnE2B9foFbm/+drfzIO7UjN+4
ETSMtixTCoGOuyPT1bhFHKrxIf3gY+OnCiwEfgitHYHq+S/PCwRMfiz9zIDfCBs5hHwcJHBxf4/+XIGGfcsh7R6DCJHzHgSihpG2
Ph6TUaTP/ZSfftGlUhtNpDYS9a8r0OXvmb3yPUDeD+K4oZyK37xHnj91D0uq6rki9LmUbGaJD6bkFEaZwuQpzDLFgKcYs
e+XdRWpyQVIoERUKskKJsFCQFkrEhYK8kBWYf0yMlRneTRMdieWPNpa/Wcv3WJk1hek38X7DVu+XrV503i9NBmaiwKqRsFc1tJIp
po/HZBTpcz/lp190qdRGE6mNRP3rCnT5e2avfA+Q94M4biin4jfvkedP3cOSqnquCH0uJZtZ4oMpOYVRpjB5CrNMMeApB
6F55d5GaXJAUSkSFgqxQIiwUpIUScaEgL2QF5h8TY2WGd9NER2L5o43lb9byPVZmTWH6TbzfsNX7ZasXnfdLk4GZKLBqJOxVDa1k
eNsEBPE2+DtLWYqBQ9oCF36ymtZdmyYqJjfYI/45cq3nS5LJ71c2fL+nib9kDC+mzMulP15nnb8gjR/QRafZJpM3
P4+GTOU0L8SZjjRuGLjOsyFk3P1re/Gj567hz9y35FWazaTbo3scT7bfFZr75rhC//iF7MFecZeTiR+KA7KDoky/8zxG0IUcy5E
Fh01wAE+Tr4O0tRgolD1g4Terae2lWbJiYqM9wr8j12qeNLnsXnXz50u6+Fv2wEL6rEz603Xm+RvyyBF9RJp9Emn
o43j4ZE7TQrzJWOOGocsMK3LWzY+WNz9a/jru3H1LfoXZbJoNuvfxRPttsZlvvivEr3/IHswVZxm5+JE4IDso+uQL/3MEbciRDD
dxBYM/Dk0UwkxR38kcjNNWliHkso3wdycIWD9mtuc3FWFGRLrF7n2qSmoyKo2qUrjDCqTOV0u1kdZLC4FXkJ
jbJqFSWVYaGPGlq0lmRUckVrLKwIqMq/0CgTmdVRiWKJGRUWfe3IqMybj2VKpuMWVdLMS9115umUuzW9nZTKUl8DFWy1GDcSipVb
Sp3ENjz8GQRzCTFnfzRCE11KWIeyyhfR7KwxUN2a25zMVZUpEvs3qeapCaj4qgapSuMcOpMpXQ7WZ2kMHgV+
bBNMiqVZZWhIU+amnRWZFRyBassrMioyj8QqNNZlVGJIgkZVdb9rciojFtPpcomY9bVUsxL3fWmqRS7tb3dVEoSH0OVLDUYt5JKV
e6NChgTwtooaMpNer8IhQ7rVcsca/ruF0ofCcM3qILzo7PpzszURmZA3VCBRQt9YtUTi35urnKjAb7S f6fuhe
+kVpLhbLfXOFFKdT/bRUG//oVmSTGJknAbdRk+P1KWpNJNkUZGd6mKLMpymyKMpuizKYo8z9VlMkusigFlVxjqyo0167PvGl9JWl
5d6okCEBvK2ihsyk16twyJButdyxhv9uofShMFyzOggvOrv+3GxNRCbkDRVIlNA3Vi2R+PfmKicq8Bu
yb6TWUqHsN1d4UQr1f1uFwb9+RSaJsUkScBs1GX6/ktZkkk1RRoa3KcpsijKbosymKLMpyvxPFWWyiyxKQSXX2KoKzbXrM29aX0n
n/qXmz
LH
Nw
Uespai8j1qaLuy8pnGkrh97t
+ph6ylqHyPGtqurHymsSRun7
LHNdd4Ju5uZtRklVkJSc
Cstc152gm7l5m1FSFSQl
iKvJLuFBU/ecyf1q
mSIq+kO0XFTx7zpz
YnbSqUkVpHS
VvitlIpiVWk
Xlg3erO
alWWDNy
= = R∗ R  0 .

Figure 1. The Picture Proof of Reflection Positivity.

9
 One then rotates part of the picture to obtain an expectation of the positive
Hilbert-space expectation of the quantity R∗ R. In the picture we see that the
positive-temperature expectation of Θ(R) ◦ R is given by another expectation of
R∗ R, and this expectation also involves the rotation of the Hamiltonian H.
There are many technical points, and in [20] you will find the complete proof
as well as the explanation of all the details. One must begin by proving that
our pictorial language is well-defined as a mathematical tool, and that it maps
onto the objects of interest in mathematical physics. For this we implemented the
parafermion algebra as an example of our model. We call the example PAPPA
(for parafermion planar para algebra).
One must establish the para-isotopy invariance of our pictures. It is necessary
to show that our expectation of pictures is a trace, and that it is positive. In this
way one can use the GNS construction to obtain a Hilbert space representation of
the pictures.
This led us to prove a variation of Jones’ famous index theorem [27] for the
quantization of the quantum dimension. Finally we had to understand how the
one-string rotation Fs of a picture (the SFT of the picture) is actually a general-
ization of the Fourier transform of the function the picture represents. One needs
to pin down the relation between the SFT and reflection invariance of H leading
to RP. It turns out that we can require that Fs (H) is positive, as it occurs in the
expectation h · i in Figure 1. It is not necessary that the Hamiltonian itself be
positive for RP to hold, but only that its SFT is positive, This can be expressed
as H having a hermitian expansion in terms of a natural basis.
Having established the RP result for parafermions, one can ask how this relates
to RP for classical or quantum spin systems. Many spin systems reduce to the
case for Majoranas, namely parafermions of degree d = 2. An extensive analysis
of the relation between RP for Majoranas and RP for classical and quantum spin
systems can be found in [18].

Acknowledgement
I am very grateful for the hospitality and working atmosphere provided by the
Research Institute for Mathematics Oberwolfach (MFO). This work was supported
in part by the Templeton Religion Trust under Grants TRT0080 and TRT0159.

References
[1] Marek Biskup, Reflection positivity and phase transitions in lat-
tice spin models, arXiv:math-ph/0610025, In:Methods of Con-
temporary Mathematical Statistical Physics, R. Kotecky (Editor),
Lecture Notes in Mathematics, 1970 (2009), 1–86, Springer Berlin Heidelberg.
[2] Stefano Chesi, Arthur Jaffe, Daniel Loss, and Fabio L. Pedrocchi, Vortex Loops and Majo-
ranas, J. Math. Phys. 54 (2013), 112203.
[3] Roland Dobrushin and Robert Minlos, Construction of a One-
Dimensional Quantum Field by Means of a Continuous Markov Field,
Functional Analysis and Applications 7 (1973), 324–425.
[4] Freeman Dyson, Elliott Lieb, and Barry Simon, Phase transitions in quantum spin systems
with isotropic and nonisotropic interactions, J. Stat. Phys. 18 (1978), 335–383.
10
 [5] Rupert L. Frank, contribution to this meeting.
[6] Jürg Fröhlich, The Reconstruction of Quantum Fields from Euclidean Greens Functions at
Arbitrary Temperatures, Helv. Phys. Acta 48(3) (1975), 355–369.
[7] , contribution to this meeting.
[8] Jürg Fröhlich, Robert Israel, Elliott Lieb, and Barry Simon, Phase Transitions
and Reflection Positivity. I. General Theory and Long Range Lattice Models,
Comm. Math. Phys., 62 (1978), 1–34.
[9] Jürg Fröhlich, Konrad Osterwalder and Erhard Seiler, On Virtual Representations of Sym-
metric Spaces and Their Analytic Continuation, Ann. of Math., 118(3) (1983) 461–489.
[10] Jürg Fröhlich, Barry Simon, and Thomas Spencer, Infrared bounds, phase transitions and
continuous symmetry breaking, Commun. Math. Phys., 50 (1976), 79–95.
[11] James Glimm and Arthur Jaffe, A Note on Reflection Positivity,
Lett. Math. Phys., 3(5) (1979), 377–378.
[12] James Glimm, Arthur Jaffe, and Thomas Spencer, The Wightman Axioms and Particle
Structure in the P (φ)2 Quantum Field Model, Ann. of Math., 100(3) (1974), 585–632.
[13] , Phase Transitions for ϕ22 Quantum Fields,
Commun. Math. Phys., 45 (1975), 203–216.
[14] Rudolf Haag, N.N. Hugenholtz, and M. Winnink, On the equilibrium states in quantum
statistical mechanics, Comm. Math. Phys., 5(3) (1967), 215–236.
[15] Arthur Jaffe, Stochastic PDE, Reflection Positivity, and Quantum Fields,
Journal of Statistical Physics, 161(1) (2015), 1–15.
[16] Arthur Jaffe, Christian D. Jakel, Roberto E. Martinez, II, Complex Classical Fields: a
Framework for Reflection Positivity, Comm. Math. Phys., 329(1) (2014), 1–28.
[17] , Complex classical fields: an example, J. Funct. Anal. 266(3) (2014), 1833–1881.
[18] Arthur Jaffe and Bas Janssens, Characterization of Reflection Positivity: Majoranas and
Spins, Commun. Math. Phys., 346(3) (2016), 1021–1050.
[19] , Reflection Positive Doubles, Jour. Funct, Anal., 272(8) (2017), 3506–3557.
[20] Arthur Jaffe and Zhengwei Liu, Planar Para Algebras and Reflection Positivity,
Commun. Math. Phys. 352 (2017), 95–133.
[21] , Mathematical Picture Language Program, PNAS, 115 (2018), 81–86.
[22] Arthur Jaffe, Zhengwei Liu, and Alex Wozniakowski, Holographic Software for Quantum
Networks, SCIENCE CHINA Mathematics, 61(4) (2018), 593–626.
[23] Constructive Simulation and Topological Design of Protocols,
New Journal of Physics, 19 (2017), 063016.
[24] Arthur Jaffe, Fabio L. Pedrocchi, Reflection Positivity for Parafermions,
Comm. Math. Phys. 337(1) (2015), 455–472.
[25] Chunlan Jiang, Zhengwei Liu, and Jinsong Wu, Noncommutative uncertainty principles,
J. Funct. Anal., 270(1) (2016), 264–311.
[26] , Block maps and Fourier analysis, SCIENCE CHINA, Mathematics, to appear.
[27] Vaughan F. R. Jones, Index for subfactors, Invent. Math., 72 (1983), 1–25.
[28] Palle Jorgensen, contribution to this meeting.
[29] Palle Jorgensen, Karl-Hermann Neeb, and Gestur Ólafsson, Re-
flection Positive Stochastic Processes Indexed by Lie Groups,
Symmetry, Integrability and Geometry: Methods and Applications (SIGMA), 12 (2016), 058, 49 pages.
[30] Res Jost, The General Theory of Quantized Fields, American Mathematical Society, Prov-
idence, RI, (1965).
[31] Abel Klein, Gaussian OS-positive processes, Z. Wahrscheinlichkeitstheorie und Verw. Gebiete 40 (1977), 115–124.
[32] Abel Klein and Larry Landau, Construction of a unique self-adjoint generator for a sym-
metric local semigroup. J. Funct. Anal., 44 (1981), 121–137.
[33] , From the Euclidean group to the Poincaré group via Osterwalder-Schrader positiv-
ity, Commun. Math. Phys., 87 (1983), 469–484.
[34] Elliott Lieb, Flux phase of the half-filled band, Phys. Rev. Lett. 73 (1994), 2158–2161.

11
[35] Zhengwei Liu, Alex Wozniakowski, and Arthur Jaffe, Quon 3D Language for Quantum
Information, PNAS, 114 (2017), 2497–2502.
[36] Roberto Longo, contribution to this meeting.
[37] Michael Miller, The Origins of Schwinger’s Euclidean Green’s Functions,
Studies in History and Philosophy of Modern Physics, 50 (2015), 5–12.
[38] Edward Nelson, Construction of Quantum Fields from Markoff Fields,
J. Funct. Anal. 12 (1973), 97–112.
[39] , The Free Markoff Field, J. Funct. Anal. 12 (1973), 211–227.
[40] , Probability Theory and Euclidean Field Theory, in Constructive Quantum Field
Theory. The 1973 “Ettore Majorana” International School of Mathematical Physics, edited
by G. Velo and A. Wightman, pp. 94–124, Springer Verlag, Berlin (1973).
[41] Konrad Osterwalder, Euclidean Green’s Functions and Wightman Distributions, in Con-
structive Quantum Field Theory. The 1973 “Ettore Majorana” International School of
Mathematical Physics, edited by G. Velo and A. Wightman, pp. 71–93, Springer Verlag,
Berlin (1973).
[42] Konrad Osterwalder and Robert Schrader, Axioms for Euclidean Green’s Functions,
Comm. Math. Phys., 31 (1973), 83–112.
[43] , Axioms for Euclidean Green’s functions II. Comm. Math. Phys. 42 (1975) 281–305.
[44] Konrad Osterwalder and Erhard Seiler, Gauge field theories on a lattice,
Ann. Phys., 110(2) (1978), 440-471.
[45] Julian Schwinger, Four-Dimensional Euclidean Formulation of Quantum Field Theory,
1958 Annual International Conference on High Energy Physics, at CERN, edited by B. Ferretti, pp.134–139.
CERN Scientific Information Service, Genève.
[46] , On the Euclidean structure of relativistic field theory,
Proceedings of the National Academy of Sciences, 44 (1958), 956–965.
[47] Kurt Symanzik, A Modified Model of Euclidean Quantum Field Theory,
Courant Institute of Mathematical Sciences Report IMM-NYU 327, 132 pages, June 1964.
[48] , Euclidean Quantum Field Theory. I. Equations for a Scalar Model,
Journal of Mathematical Physics, 7 (1966), 510–525.
[49] , Euclidean Quantum Field Theory, in Local Quantum Theory, Proceedings of the
1968 Varenna School of Physics, R. Jost, Editor, Academic Press, New York, 1969.
[50] Masamichi Takesaki, Tomita’s Theory of Modular Hilbert Algebras and Its Applications,
Lecture Notes in Mathematics, Vol. 128. Springer-Verlag, Berlin, Heidelberg, and New
York (1970).
[51] Gian-Carlo Wick, Properties of Bethe-Salpeter Wave Functions,
Physical Review, 96 (1954), 1124–1134.
[52] David V. Widder, Necessary and sufficient conditions for the representation of a function
by a doubly infinite laplace integral, Bull. Amer. Math. Soc., 40(4) (1934), 321–326.
[53] Arthur Wightman, Quantum field theory in terms of vacuum expectation values,
Physical Review, 101 (1956), 860–866.

12