Causal Processes in C*-Algebraic Setting


In this paper, we attempt to explicate Salmon’s idea of a causal process, as defined in terms of the mark method, in the context of C*-dynamical systems. We prove two propositions, one establishing mark manifestation infinitely many times along a given interval of the process, and, a second one, which establishes continuous manifestation of mark with the exception of a countable number of isolated points. Furthermore, we discuss how these results can be implemented in the context of the Haag–Araki theories of relativistic quantum fields on Minkowski spacetime.

This is a preview of subscription content, access via your institution.


  1. 1.

    Gleason’s theorem and its consequences for a probabilistic interpretation of a quantum theory and the preparability of a state by means of physically realizable local operations in an open bounded region of Minkowski spacetime are some reasons for restricting physically admissible states to normal states. See [7].

  2. 2.

    For a finite state space \({\mathscr {X}}=\left\{ x_{1},\ldots ,x_{n}\right\}\), the algebra of observables is \(C({\mathscr {X}})=\left\{ f:f:{\mathscr {X}}\rightarrow {\mathbb {C}}\right\}\) and let \(T:C({\mathscr {X}})\rightarrow C({\mathscr {X}})\) be a classical operation, such that \(g=T(f):\left[ \begin{array}{c} g(x_{1})\\ \vdots \\ g(x_{n}) \end{array}\right]\) =\(\left[ \begin{array}{ccc} T_{x_{1}x_{1}} &{} \ldots &{} T_{x_{1}x_{n}}\\ \vdots &{} \ddots &{} \vdots \\ T_{x_{n}x_{1}} &{} \cdots &{} T_{x_{n}x_{1}} \end{array}\right]\) \(\left[ \begin{array}{c} f(x_{1})\\ \vdots \\ f(x_{n}) \end{array}\right]\) for every \(f\in C({\mathscr {X}})\).

  3. 3.

    That for several Haag–Araki theories of physical significance the local algebras \({\mathcal {R}}\left( O\right)\) of finite regions of Minkowski spacetime are type \(III_{1}\) factors, see [11, Thm. 1.3.12, pp. 35 and 254]), and that the global algebra \({\mathcal {R}}\) of a system may be a type \(I_{\infty }\) von Neumann factor if the system does not have superselection sectors, see [11, p. 24]).

  4. 4.

    For a brief and informative presentation of the Lüders rule, see [12]. The original article in German is [13] and an English translation is [14].

  5. 5.

    More examples are provided by quantum channels in quantum information theory. For instance, the amplitude-damping channel provides a simple model for the spontaneous emission of a photon in the transition of a two-level atom from its excited to its ground state. It is defined in terms of the map \(T:{{\mathcal {B}}}({\mathbb {C}}^{2})\rightarrow \mathcal {{\mathcal {B}}}({\mathbb {C}}^{2}):X\mapsto M_{o}^{*}XM_{o}+M_{1}^{*}XM_{1}\) where \(M_{0}=\left[ \begin{array}{cc} 1 &{} 0\\ 0 &{} \sqrt{1-p} \end{array}\right]\), \(M_{1}=\left[ \begin{array}{cc} 0 &{} \sqrt{p}\\ 0 &{} 0 \end{array}\right]\) and p is the probability of decay.

  6. 6.

    If we do not make this assumption and we consider any real number \(t_{1}\ne 0\), the interval over which the mark is transmitted is \((a,b)\cup \left\{ t_{1}\right\}\), where \(a=min\left\{ 0,t_{1}\right\}\) and \(b=max\left\{ 0,t_{1}\right\} .\) Thus, in Defs. 4 and 5 there is no hidden preference about the direction of transmission of the mark along a process, in conformity with Salmon’s desideratum of providing definitions of the fundamental causal notions that are not committed to a preferred time direction.

  7. 7.

    See “Appendix

  8. 8.

    In 1964, Haag and Kastler in a well-known paper provided an axiomatic formulation of quantum theory of fields in terms of nets of abstract C*-algebras of local observables instead of quantum fields that, henceforth, is known as the Haag–Kastler formulation [21]. The same year, Araki published two papers [22]  and [23], in which he used nets of von Neumann algebras of bounded operators on a Hilbert space to axiomatize quantum field theory using analogous axioms to the Haag–Kastler ones. This formulation is known as the Haag–Araki approach to quantum field theory.

  9. 9.

    The discussion in this Valente’s paper refers to almost local operators which are constructed as analytic elements for the generators of the group of translations of Minkowski spacetime.

  10. 10.

    Private communication.


  1. 1.

    Reichenbach, H.: The Philosophy of Space and Time (Reichenbach , M., Freund, J., trans.). Dover, New York (1928 [1958])

  2. 2.

    Russell, B.: Human Knowledge: Its Scope and Limits. George Allen and Unwin, London (1923)

    Google Scholar 

  3. 3.

    Salmon, W.: Scientific Explanation and the Causal Structure of the World. Princeton University Press, Princeton (1984)

    Google Scholar 

  4. 4.

    Salmon, W.: Causality and Explanation. Oxford University Press, Oxford (1998)

    Google Scholar 

  5. 5.

    Dowe, P.: Wesley Salmon’s process theory of causality and the conserved quantity theory. Philos. Sci. 59, 195–216 (1992)

    Article  Google Scholar 

  6. 6.

    Dowe, P.: Physical Causation. Cambridge University Press, Cambridge (2000)

    Google Scholar 

  7. 7.

    Ruetsche, L., Earman, J.: Probabilities in quantum field theory and quantum statistical mechanics. In: Beisbart, C., Hartmann, S. (eds) Probabilities in Physics, pp. 263–290. Oxford University Press, Oxford (2011)

    Google Scholar 

  8. 8.

    Salmon, W.: Causality and explanation: a reply to two critiques. Philos. Sci. 64, 461–477 (1997)

    Article  Google Scholar 

  9. 9.

    Hellwig, K.E., Kraus, K.: Operations and measurements. II. Commun. Math. Phys. 16, 142–147 (1970)

    ADS  MathSciNet  Article  Google Scholar 

  10. 10.

    Takesaki, M.: Theory of Operator Algebras I. Springer, Berlin (1979 [2000])

  11. 11.

    Horuzhy, S.S.: Introduction to Algebraic Quantum Field Theory. Kluwer Academic, Dordrecht (1990)

    Google Scholar 

  12. 12.

    Busch, P., Lahti, P.: Lüders rule. In: Greenberger, D., et al. (eds) Compendium of Quantum Physics: Concepts, Experiments. History and Philosophy. Springer, Berlin (2009)

    Google Scholar 

  13. 13.

    Lüders, G.: Über die Zustandsnderung durch den Meßprozeß. Ann. Phys. (Leipz.) 8, 322–328 (1951)

    MATH  Google Scholar 

  14. 14.

    Lüders, G.: Concerning the state-change due to the measurement process. Ann. Phys. (Leipz.) 15, 663–670 (1951 [2006])

  15. 15.

    Clifton, R., Halvorson, H.: Entanglement and open systems in algebraic quantum field theory. Stud. Hist. Philos. Mod. Phys. 32, 1–31 (2001)

    MathSciNet  Article  Google Scholar 

  16. 16.

    Clifton, R., Bub, J., Halvorson, H.: Characterizing quantum theory in terms of information-theoretic constraints. Found. Phys. 33, 1561–1591 (2003)

    MathSciNet  Article  Google Scholar 

  17. 17.

    Borchers, H.J.: A remark on a theorem of B. Misra. Commun. Math. Phys. 4, 315–323 (1967)

    ADS  MathSciNet  Article  Google Scholar 

  18. 18.

    Earman, J., Valente, G.: Relativistic causality in algebraic quantum field theory. Int. Stud. Philos. Sci. 28, 1–48 (2014)

    MathSciNet  Article  Google Scholar 

  19. 19.

    Lang, S.: Complex Analysis. Springer, New York (1999)

    Google Scholar 

  20. 20.

    Haag, R.: Local Quantum Physics: Fields, Particles. Algebras. Springer, Berlin (1996)

    Google Scholar 

  21. 21.

    Haag, R., Kastler, D.: An algebraic approach to quantum field theory. J. Math. Phys. 5(1), 848–862 (1964)

    ADS  MathSciNet  Article  Google Scholar 

  22. 22.

    Araki, H.: Von Neumann algebras of local observables of free scalar field. J. Math. Phys. 5(1), 1–13 (1964)

    ADS  MathSciNet  Article  Google Scholar 

  23. 23.

    Araki, H.: On the algebras of all local observables. Prog. Theor. Phys. 32(5), 844–854 (1964a)

    ADS  MathSciNet  Article  Google Scholar 

  24. 24.

    Reeh, H., Schlieder, S.: Bemerkungen zur Unitrquivalenz von Lorentzinvarienten Feldern. Nuovo Cimento 22, 1051–1068 (1961)

    Article  Google Scholar 

  25. 25.

    Valente, G.: Restoring particle phenomenology. Stud. Hist. Philos. Mod. Phys. 51, 97–103 (2015)

    MathSciNet  Article  Google Scholar 

  26. 26.

    Arageorgis, A., Stergiou, C.: On particle phenomenology without particle ontology: how much local is almost local? Found. Phys. 43, 969–977 (2013)

    ADS  MathSciNet  Article  Google Scholar 

  27. 27.

    Steinmann, O.: Particle localization in field theory. Commun. Math. Phys. 7, 112–137 (1968)

    ADS  MathSciNet  Article  Google Scholar 

  28. 28.

    Bratteli, O., Robinson, D.W.: Operator Algebras and Quantum Statistical Mechanics 1: C*- and W*-Algebras, Symmetry Groups. Decomposition of States. Springer, Berlin (1987)

    Google Scholar 

Download references


The author expresses his gratitude to Professor Dr. K. Fredenhagen for his substantial help. Also, the author would like thank the two anonymous referees for contributing with their comments to the improvement of the manuscript.

Author information



Corresponding author

Correspondence to Chrysovalantis Stergiou.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendix: On Analytic Elements for the One-Parameter Group of Time Translations in Local Quantum Physics

Appendix: On Analytic Elements for the One-Parameter Group of Time Translations in Local Quantum Physics

Given a \(\sigma (X;F)\)-continuous group of isometries of a complex Banach space X, an element \(A\in X\) is said to be analytic for \(\tau\) ( \(\tau\)-analytic) if there exist a strip

$$I_{\lambda }=\left\{ z\in {\mathbb {C}}:\mid Imz\mid <\lambda \right\}$$

in \({\mathbb {C}}\), and a function \(f_{A}:I_{\lambda }\rightarrow X\) such that

(i) \(f_{A}(t)=\tau _{t}(A)\) for \(t\in {\mathbb {R}},\)

(ii) \(z\mapsto \eta (f_{A}(z))\) is analytic for all \(\eta \in F\) and \(z\in I_{\lambda }\).

On these conditions, \(f_{A}(z)=\tau _{z}(A),\,z\in I_{\lambda }\) is an X-valued analytic function; if \(\lambda =\infty\) then\(f_{A}\) is entirely analytic [28, Def. 2.5.20, p. 99]. Moreover, condition (ii) is equivalent to the existence of the limit

$$\underset{h\rightarrow 0}{lim}\,\frac{f_{A}(z+h)-f_{A}(z)}{h}$$

in norm in X for \(z\in I_{\lambda }\) as h tends to zero in \({\mathbb {C}}\), [28, Prop. 2.5.21, p. 99].

The groups of isometries we consider in this paper are \(\sigma ({\mathcal {A}},{\mathcal {A}}^{*})\)-continuous i.e. for all \(A\in {\mathcal {A}},\,t\mapsto \eta (\tau _{t}(A))\) is continuous for every \(\eta \in {\mathcal {A}}^{*},\)which amounts to the requirement that \(t\mapsto \tau _{t}(A)\) be continuous in norm for every \(A\in \mathcal {A}\) where \({\mathcal {A}}\) is a von Neumann algebra.


The sum and the product of two analytic elements for a group of automorphisms \(\tau\) in a von Neumann algebra \({\mathcal {A}}\) is a \(\tau\)-analytic element.


Consider two \(\tau\)-analytic elements \(A,B\in {\mathcal {A}}\) and let \(f_{A}(z)=\tau _{z}(A),\,z\in I_{\lambda _{1}}\), \(f_{B}(z)=\tau _{z}(B),\,z\in I_{\lambda _{2}}\) for \(I_{\lambda _{i}}=\left\{ z\in {\mathbb {C}}:\mid Imz\mid <\lambda _{i}\right\} ,i=1,2\) be the two corresponding \({\mathcal {A}}\)-valued functions that satisfy conditions (i) and (ii). To show that the product AB is \(\tau\)-analytic we should prove that the \({\mathcal {A}}\)-valued function \(f_{AB}:I_{\lambda }\rightarrow {\mathcal {A}},\;f_{AB}(z)=f_{A}(z)f_{B}(z)\) which extends the function \(f_{AB}(t)=f_{A}(t)f_{B}(t)=\tau _{t}(A)\tau _{t}(B)=\tau _{t}(AB)\) for \(t\in {\mathbb {R}},\)satisfies the condition of analyticity:

$$\underset{h\rightarrow 0}{lim}\,\frac{f_{AB}(z+h)-f_{AB}(z)}{h}$$

in \(I_{\lambda }=\left\{ z\in {\mathbb {C}}:\mid Imz\mid <\lambda \right\} ,\lambda =min\left\{ \lambda _{1},\lambda _{2}\right\}\). But

$$\begin{aligned}&\frac{f_{AB}(z+h)-f_{AB}(z)}{h}=\frac{f_{A}(z+h)f_{B}(z+h)-f_{A}(z)f_{B}(z)}{h}\\&\quad =\frac{f_{A}(z+h)-f_{A}(z)}{h}f_{B}(z+h)+f_{A}(z)\frac{f_{B}(z+h)-f_{B}(z)}{h}. \end{aligned}$$

Since \(f_{A}\)and \(f_{B}\) satisfy the analyticity condition, which implies also the continuity of \(f_{A}\)and \(f_{B}\), each factor in the last expression in the above equality has a well-defined limit for \(h\rightarrow 0\) and the analyticity condition for \(f_{AB}\) is satisfied, for every \(z\in I_{\lambda }\). Hence, AB is a \(\tau\)-analytic element of \({\mathcal {A}}.\)

The case of the sum of two \(\tau\)-analytic elements can be proven along the same line of reasoning by a simple direct calculation. \(\square\)

To obtain a family of analytic elements for the group of time translations one may begin with any element localized in some open bounded region O of Minkowski spacetime, \(A\in {\mathcal {R}}(O)\), and smear its time translation over \({\mathbb {R}}\) with a Gaussian function depending on a parameter \(n\in {\mathbb {N}}.\) Namely,

$$A_{n}=\sqrt{\frac{n}{\pi }}\int _{-\infty }^{+\infty }e^{-nt^{2}}\tau _{t}(A)dt,\quad n\in {\mathbb {N}}.$$

Each \(A_{n}\) is an entire analytic element for \(\tau\) and the family \(\left\{ A_{n}\right\} _{n\in {\mathbb {N}}}\) converges to A in the \(\sigma (X;F)\) topology. Moreover, in [28, Cor. 2.5.23, p. 101] it is shown that the set of entire analytic elements form \(\sigma ({\mathcal {R}};{\mathcal {R}}^{*})\)- continuous group of isometries form a norm-dense subset of \({\mathcal {R}}.\)


(Fredenhagen, 2019)Footnote 10 In a Haag–Araki theory of relativistic quantum fields on Minkowski spacetime, no non-trivial analytic elements for time translations can be localized in open bounded regions of Minkowski spacetime.


Let A be analytic for time translations element which is also localized in a bounded region O. Let B be localized in the spacelike complement of an \(\varepsilon\)-neighbourhood of O,  for some \(\varepsilon >0\). Then the commutator,

$$\left[ \tau _{t}(A),B\right]$$

vanishes in an open interval of t. For \(\left| t\right| <\varepsilon\) we have

$$\left\langle B^{*}\varOmega ,\tau _{t}(A)\varOmega \right\rangle =\left\langle \tau _{t}(A^{*})\varOmega ,B\varOmega \right\rangle ,$$

where \(\varOmega \in {\mathcal {H}}\) is the vacuum vector.

By the spectrum condition, which demands the generator of the time translations to be positive, the term on the left hand side defines a function that is analytic in the upper halfplane while the term on the right hand side, an analytic function in the lower halfplane. By assumption, they are also analytic in a strip around the real axis. Since these analytic functions coincide in an open interval of the real axis, we obtain an entire bounded analytic function which, therefore, is constant.

By the Reeh–Schlieder Theorem, these observables B generate from the vacuum \(\varOmega\) a dense set of vectors in the Hilbert space \({\mathcal {H}}\). Thus, \(A\varOmega\) is invariant under time evolution and, due to the uniqueness of the vacuum,

$$A\varOmega =\left\langle \varOmega ,A\varOmega \right\rangle \varOmega .$$

Thereby, we get the result that an analytic local observable is a multiple of the identity of the algebra. \(\square\)

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Stergiou, C. Causal Processes in C*-Algebraic Setting. Found Phys 51, 5 (2021).

Download citation


  • Causation
  • Process theories
  • C*-dynamical systems
  • Local quantum physics