Skip to main content
Log in

Quantum states: an analysis via the orthogonality relation

  • Original Research
  • Published:
Synthese Aims and scope Submit manuscript

Abstract

From the Hilbert space formalism we note that five simple conditions are satisfied by the orthogonality relation between the (pure) states of a quantum system. We argue, by proving a mathematical theorem, that they capture the essentials of this relation. Based on this, we investigate the rationale behind these conditions in the form of six physical hypotheses. Along the way, we reveal an implicit theoretical assumption in theories of physics and prove a theorem which formalizes the idea that the Superposition Principle makes quantum physics different from classical physics. The work follows the paradigm of mathematical foundations of quantum theory, which I will argue by methodological reflection that it exemplifies a formal approach to analysing concepts in theories.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others

Notes

  1. The term ‘(testable) property’ in quantum logic means the physical concept represented by subspaces of Hilbert spaces over \(\mathbb {C}\) according to quantum theory. It is called ‘experimental proposition’ in Birkhoff and von Neumann (1936).

  2. This acronym is not common in the literature. It may only be used in this paper, because the full term is a bit long.

  3. \(\mathbb {C}\) denotes the field of complex numbers.

  4. The way how a Hilbert space is used to model a quantum system is worth attention. The reason is that differential manifolds in classical physics and Hilbert spaces in quantum physics can both be considered as generalizations or enrichments of vector spaces. However, in classical physics each state is represented by a unique vector, while in quantum physics a state can be represented by two different vectors.

  5. Here the word ‘simple’ is used in an intuitive sense and mainly concerns the number and the complexity of primitives.

  6. There are different notions of discrimination between states in physics. For example, two are discussed in Zhang et al. (2006), and one of them called perfect discrimination is the one used here.

  7. Note that this point is non-trivial. For example, if we start from the concept of mixed states, it is not clear from the literature that a theory with similar nice features can be built.

  8. In this paper, all vector spaces, including Hilbert spaces, are assumed not to be \(\{ \mathbf {0} \}\).

  9. In this paper, a subspace of \(\mathcal {H}\) means a closed linear subspace, i.e. it forms a Hilbert space with the addition, the scalar multiplication and the inner product inherited from \(\mathcal {H}\).

  10. In many textbooks, e.g. Shankar (2008), it is usually written that the states of the system are modelled by unit vectors of \(\mathcal {H}\). However, according to the measurement postulate, in predicting experimental results there is no difference between a unit vector \(\mathbf {v}\) and a non-zero scalar multiplication \(c \mathbf {v}\) of it. Hence the statement here is more precise.

  11. In fact, there is a minor conventional difference: an inner product is conjugate on the first argument, while an Hermitian form is conjugate on the second argument.

  12. Here the term ‘Hilbertian geometry’ is used in the sense of Definition 14.5.4 in Faure and Frölicher (2000) and Definition 2.4 in Zhong (2018).

  13. The term ‘class function’ emphasizes that \(\mathbf {G}\) is defined on a proper class. Though it behaves like a function, \(\mathbf {G}\) is not a set and thus is not a function in the sense of \(\mathbf {ZFC}\).

  14. Note that what is called an ‘arguesian irreducible Hilbertian geometry’ in this paper is called an ‘arguesian Hilbert geometry’ in Stubbe and van Steirteghem (2007).

  15. \(\mathbb {H}\) denotes the division ring of quaternions.

  16. f[s] is the image of s under f. It makes sense, for f is a function from V to V and \(s \subseteq V\).

  17. This means that there is a division ring isomorphism \(\sigma \) on \(\mathcal {K}\) such that, for any \(\mathbf {u} , \mathbf {v} \in V\) and x in \(\mathcal {K}\), \(f ( \mathbf {u} + \mathbf {v} ) = f (\mathbf {u}) + f (\mathbf {v})\) and \(f (x \mathbf {u}) = \sigma (x) f (\mathbf {u})\). (Definition 6.6.10 in Faure and Frölicher (2000))

  18. This means that, for each \(t \in P_{i^*}\), a measurement of the observable modelled by \(\{ P_i \mid i \in I \}\) can discriminate between s and t.

References

  • Abramsky, S., & Coecke, B. (2004). A categorical semantics of quantum protocols. In Proceedings of the 19th IEEE conference on Logic in Computer Science (LiCS’04), pages 415–425. IEEE Press

  • Aerts, D. (2009). Quantum axiomatics. In K. Engesser, D. M. Gabbay, & D. Lehmann (Eds.), Handbook of quantum logic and quantum structures: Quantum logic (pp. 79–126). Elsevier B.V.

  • Aerts, D., & van Steirteghem, B. (2000). Quantum axiomatics and a theorem of M.P. Solèr. International Journal of Theoretical Physics, 39(3), 497–502.

    Article  Google Scholar 

  • Amemiya, I., & Araki, H. (1966). A remark on Piron’s paper. Publications of the Research Institute for Mathematical Sciences, 2(3), 423–427.

  • Baltag, A., & Smets, S. (2005). Complete axiomatizations for quantum actions. International Journal of Theoretical Physics, 44(12), 2267–2282.

    Article  Google Scholar 

  • Beltrametti, E., & Cassinelli, G. (1981). The logic of quantum mechanics. Encyclopedia of mathematics and its applications. Addison-Wesley Publishing Company.

  • Benthem van, J., & Minică, S. (2012). Toward a dynamic logic of questions. Journal of Philosophical Logic, 41, 633–669.

  • Berberian, S. K. (1961). Introduction to Hilbert space. Oxford University Press.

  • Beutelspacher, A., & Rosenbaum, U. (1998). Projective geometry: From foundations to applications. Cambridge University Press.

  • Birkhoff, G., & von Neumann, J. (1936). The logic of quantum mechanics. The Annals of Mathematics, 37, 823–843.

    Article  Google Scholar 

  • Bub, J. (1982). Quantum logic, conditional probability, and interference. Philosophy of Science, 49(3), 402–421.

    Article  Google Scholar 

  • Dalla Chiara, M.L, Giuntini, R., & Greechie, R. (2004). Reasoning in quantum theory: Sharp and unsharp quantum logics, volume 22 of Trends in logic. Kluwer Acadamic Press.

  • Coecke, B., & Kissinger, A. (2017). Picturing quantum processes: A first course in quantum theory and diagrammatic reasoning. Cambridge University Press.

  • D’Ariano, G. M., Chiribella, G., & Perinotti, P. (2017). Quantum theory from first principles: An informational approach. Cambridge University Press.

  • Dishkant, H. (1972). Semantics of the minimal logic of quantum mechanics. Studia Logica, 30(1), 23–30.

    Article  Google Scholar 

  • Faure, C.-A., & Frölicher, A. (2000). Modern projective geometry. Springer.

  • Foulis, D. J., & Randall, C. H. (1971). Lexicographic orthogonality. Journal of Combinatorial Theory, Series A, 11(2), 157–162.

    Article  Google Scholar 

  • Goldblatt, R. I. (1974). Semantic analysis of orthologic. Journal of Philosophical Logic, 3, 19–35.

    Article  Google Scholar 

  • Goldblatt, R. I. (1975). The Stone space of an ortholattice. Bulletin of the London Mathematical Society, 7(1), 45–48.

    Article  Google Scholar 

  • Groenendijk, J., & Stokhof, M. (2011). Questions. In J. van Benthem & A. ter Meulen (Eds.), Handbook of logic and language (2nd ed., pp. 1059–1131). Elsevier B.V.

  • Hall, B. C. (2013). Quantum theory for mathematicians. Springer.

  • Halmos, P. R. (1951). Introduction to Hilbert space and the theory of spectral multiplicity. Chelsea Publishing Company.

  • Hedlíková, J., & Pulmannová, S. (1991). Orthogonality spaces and atomistic orthocomplemented lattices. Czechoslovak Mathematical Journal, 41, 8–23.

    Article  Google Scholar 

  • Holland, S. S. (1995). Orthomodularity in infinite dimensions: A theorem of M. Solèr. Bull. Amer. Math. Soc., 32, 205–234.

    Article  Google Scholar 

  • Kadison, R.V., & Ringrose, J.R. (1997). Fundamentals of the theory of operator algebras, Volumn I: Elementary theory. American Mathematical Society

  • Kalmbach, G. (1983). Orthomodular lattices. Acadamic Press.

  • Mayet, R. (1998). Some characterizations of the underlying division ring of a Hilbert lattice by automorphisms. International Journal of Theoretical Physics, 37, 109–114.

    Article  Google Scholar 

  • Mielnik, B. (1968). Geometry of quantum states. Communications in Mathematical Physics, 9(1), 55–80.

    Article  Google Scholar 

  • Mittelstaedt, P. (1978). Quantum logic. D. Reidel Publishing Company.

  • Mittelstaedt, P. (2013). Rational Reconstructions of Modern Physics (2nd enlarged). Springer.

  • Moore, D. J. (1995). Categories of representations of physical systems. Helvetica Physica Acta, 68, 658–678.

    Google Scholar 

  • Piron, C. (1976). Foundations of Quantum Physics. Reading: W.A. Benjamin Inc.

  • Shankar, R. (2008). Principles of quantum mechanics (2nd ed.). Plenum Press.

  • Solèr, M. P. (1995). Characterization of Hilbert spaces with orthomodularity spaces. Communications in Algebra, 23, 219–243.

    Article  Google Scholar 

  • Stubbe, I., & van Steirteghem, B. (2007). Propositional systems, Hilbert lattices and generalized Hilbert spaces. In K. Engesser, D. M. Gabbay, & D. Lehmann (Eds.), Handbook of quantum logic and quantum structures: Quantum structures (pp. 477–523). Elsevier B.V.

  • von Neumann, J. (1932). Mathematische Grundlagen der Quantenmechanik. Julius Springer.

  • Wigner, E.P. (1959). Group Theory and Its Application to the Quantum Mechanics of Atomic Spectra. Academic Press Inc., expanded and improved edition, translated from the German by J.J. Griffin.

  • Zabey, P. (1975). Reconstruction theorems in quantum mechanics. Foundations of Physics, 5(2), 323–342.

    Article  Google Scholar 

  • Zhang, C., Feng, Y., & Ying, M. (2006). Unambiguous discrimination of mixed quantum states. Physics Letters A, 353(4), 300–306.

    Article  Google Scholar 

  • Zhong, S. (2015) Orthogonality and Quantum Geometry: Towards a Relational Reconstruction of Quantum Theory. PhD thesis, University of Amsterdam, http://www.illc.uva.nl/Research/Publications/Dissertations/DS-2015-03.text.pdf.

  • Zhong, S. (2017). A formal state-property duality in quantum logic. Studies in Logic, 10(2), 112–133.

    Google Scholar 

  • Zhong, S. (2018). Correspondence between Kripke frames and projective geometries. Studia Logica, 106, 167–189.

    Article  Google Scholar 

Download references

Acknowledgements

All technical results in Sections 2 and A.1 are from my PhD thesis Zhong (2015) finished at the Institute for Logic, Language and Computation (ILLC), University of Amsterdam, in 2015. I am very grateful to my supervisors, Prof. Johan van Benthem, Dr. Alexandru Baltag and Prof. Sonja Smets, for their supervision and the detailed comments on my thesis which are invaluable in the writing of this paper. I am also very grateful to the members in my PhD committee, Dr. Nick Bezhanishvili, Prof. Robert Goldblatt, Prof. John Harding, Prof. Yde Venema, Prof. Ronald de Wolf and Prof. Mingsheng Ying, for their comments on my thesis which are very helpful in writing this paper. The results were also presented in many workshops and seminars, and I am very grateful to the audiences for the interesting discussion. I also thank very much the three anonymous reviewers for their helpful comments. Finally, my PhD research was funded by China Scholarship Council (CSC), and the writing of this paper is supported by the National Social Science Fund of China (No. 20CZX048).

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Shengyang Zhong.

Additional information

Publisher's Note

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

Appendix

Appendix

In this appendix, I will develop the technicalities used in the proof of Proposition 4.5 in Sect. 4.3.

We fix a Kripke frame \(\mathfrak {F} = ( \Sigma , {\rightarrow } )\) satisfying Reflexivity and Symmetry throughout this appendix for convenience.

Note that Items 1 to 5 in Lemma 2.8 only need Reflexivity and Symmetry in their proofs, so they hold in \(\mathfrak {F}\).

1.1 Subframe

In this subsection, we introduce and discuss the notion of a subframe.

Definition A.1

A subframe of \(\mathfrak {F}\) is an ordered pair \((P , {\rightarrow _P})\) such that

  1. 1.

    \(P \subseteq \Sigma \) is non-empty and closed in \(\mathfrak {F}\);

  2. 2.

    \({\rightarrow _P}\) is the restriction of \({\rightarrow }\) to P, i.e. \({\rightarrow _P} = {\rightarrow } \cap (P \times P)\).

Remark A.2

Let \((P , {\rightarrow _P})\) be a subframe of \(\mathfrak {F}\).

  1. 1.

    For any \(s,t \in P\), \(s \rightarrow t\) if and only if \(s \rightarrow _P t\).

  2. 2.

    For any \(A \subseteq P\) and \(s,t \in P\), s and t are indistinguishable with respect to A in \((P , {\rightarrow _P})\) if and only if they are indistinguishable with respect to A in \(\mathfrak {F}\).

  3. 3.

    \((P , {\rightarrow _P})\) is a Kripke frame satisfying Reflexivity and Symmetry, so the results in this appendix apply to it as well.

Given this remark, in the following, I denote the restriction of \({\rightarrow }\) in \(\mathfrak {F}\) to P by \({\rightarrow }\) instead of \({\rightarrow _P}\). I also use the same symbol \({\approx _A}\) indexed by \(A \subseteq P\) to denote the indistinguishability relation with respect to A in \((P , {\rightarrow })\) and that in \(\mathfrak {F}\). On the contrary, I denote the orthocomplement of \(A \subseteq P\) in \((P , {\rightarrow })\) by \({\sim }_P A\), because the orthocomplements in the subframe are different from those in \(\mathfrak {F}\).

I start from some simple facts about the orthocomplements and the closed subsets in a subframe.

Lemma A.3

Let \((P , {\rightarrow })\) be a subframe of \(\mathfrak {F}\).

  1. 1.

    \({\sim }_P A = {\sim }A \cap P\), for every \(A \subseteq P\).

  2. 2.

    For each \(A \subseteq P\), if \({\sim }_P {\sim }_P A = A\), then \({\sim }{\sim }A = A\).

  3. 3.

    If \(\mathfrak {F}\) satisfies Representation, then, for each \(A \subseteq P\), \({\sim }{\sim }A = A\) implies that \({\sim }_P {\sim }_P A = A\).

Proof

For Item 1: Easy verification.

For Item 2: Assume that \({\sim }_P {\sim }_P A = A\). By Lemma 2.8\(A \subseteq {\sim }{\sim }A\). It remains to show that \({\sim }{\sim }A \subseteq A\). Assume that \(s \not \in A\). By the assumption \(s \not \in {\sim }_P {\sim }_P A\). Then there is a \(t \in {\sim }_P A\) such that \(s \rightarrow t\). By 1 \(t \in {\sim }A \cap P\). Hence \(t \in {\sim }A\) is such that \(s \rightarrow t\), and thus \(s \not \in {\sim }{\sim }A\).

For Item 3: Assume that \(\mathfrak {F}\) satisfies Representation and \({\sim }{\sim }A = A\). By 1 \({\sim }_P {\sim }_P A = {\sim }({\sim }A \cap P) \cap P\), so it amounts to show that \({\sim }({\sim }A \cap P) \cap P = A\).

On the one hand, since \(A \subseteq P\) and \(A = {\sim }{\sim }A \subseteq {\sim }( {\sim }A \cap P )\) by Lemma 2.8, \(A \subseteq {\sim }({\sim }A \cap P) \cap P\).

On the other hand, let \(s \in {\sim }({\sim }A \cap P) \cap P\) be arbitrary. Then \(s \in P\) and \(s \in {\sim }({\sim }A \cap P)\). To show that \(s \in A\), by the assumption it suffices to show that \(s \in {\sim }{\sim }A\). Hence we let \(t \in {\sim }A\) be arbitrary and try to show that \(s \not \rightarrow t\). Consider two cases.

  • Case 1: \(t \in {\sim }P\).

    Since \(s \in P\), it follows that \(s \not \rightarrow t\).

  • Case 2: \(t \not \in {\sim }P\).

    By Representation there is a \(t_\parallel \in P\) such that \(t \approx _P t_\parallel \). Since \(A \subseteq P\) and \(t \in {\sim }A\), \(t_\parallel \in {\sim }A\). Hence \(t_\parallel \in P \cap {\sim }A\). Since \(s \in {\sim }({\sim }A \cap P)\), \(t_\parallel \not \rightarrow s\). Since \(s \in P\), by \(t \approx _P t_\parallel \) we have \(t \not \rightarrow s\), so \(s \not \rightarrow t\).

\(\square \)

Proposition A.4

Let \((P , {\rightarrow })\) be a subframe of \(\mathfrak {F}\). If \(\mathfrak {F}\) satisfies both Separation and Representation, then so is \((P , {\rightarrow })\).

Proof

For Representation, let \(A \subseteq P\) and \(s \in P\) be such that \(A = {\sim }_P {\sim }_P A\) and \(s\not \in {\sim }_P A\). By Lemma A.3\(A = {\sim }{\sim }A\). Moreover, since \(s \not \in {\sim }_P A\) and \(s \in P\), by Lemma A.3\(s \not \in {\sim }A\). By Representation there is an \(s' \in A\) such that \(s \approx _A s'\).

For Separation, assume that \(s,t \in P\) are such that \(s \ne t\). Since \(\mathfrak {F}\) satisfies Separation, there is a \(w \in \Sigma \) such that \(w \rightarrow s\) and \(w \not \rightarrow t\). Since \(w \rightarrow s\), \(w \not \in {\sim }P\). By Representation there is a \(w' \in P\) such that \(w \approx _P w'\). Hence \(w' \in P\) is such that \(w' \rightarrow s\) and \(w' \not \rightarrow t\). \(\square \)

1.2 Maximal Opposite Family

Here we prove some useful results about maximal opposite families.

Corollary A.5

Let \(\{ P_i \mid i \in I \}\) be a maximal opposite family in \(\mathfrak {F}\). \(\emptyset = \bigcap \{ {\sim }P_i \mid i \in I \}\).

Proof

If I is a singleton, the result holds by Lemma 2.8 and Remark 3.2. In the following we focus on the case when I has at least two elements.

Take an \(i^* \in I\). By Lemma 3.4\(P_{i^*} = \bigcap \big \{ {\sim }P_i \mid i \in I \setminus \{ i^* \} \big \}\). Hence \(\bigcap \{ {\sim }P_i \mid i \in I \} = {\sim }P_{i^*} \cap \bigcap \big \{ {\sim }P_i \mid i \in I \setminus \{ i^* \} \big \} = {\sim }P_{i^*} \cap P_{i^*} = \emptyset \). \(\square \)

Lemma A.6

If \(\mathfrak {F}\) satisfies Separation and Representation, then, for any closed \(P \varsubsetneqq \Sigma \) and \(s \not \in {\sim }P\), there are \(s_\parallel \in P\) and \(s_\bot \in {\sim }P\) such that \(s_\parallel \in {\sim }{\sim }\{ s , s_\bot \}\).

Proof

Assume that \(P \varsubsetneqq \Sigma \) is closed and \(s \not \in {\sim }P\). Note that \({\sim }P \ne \emptyset \) for \(P \ne \Sigma \). If \(s \in P\), then by just letting \(s_\parallel = s\) and picking an arbitrary \(s_\bot \in {\sim }P\) we will have \(s_\parallel \in {\sim }{\sim }\{ s , s_\bot \}\). In the following, we focus on the case when \(s \not \in P\).

Since P is closed, \(s \not \in {\sim }{\sim }P\). By Representation there is an \(s_\bot \in {\sim }P\) such that \(s \approx _{{\sim }P} s_\bot \). Since \(s \not \in {\sim }P\), \(s \ne s_\bot \), so \(s \not \in {\sim }{\sim }\{ s_\bot \}\) by Separation and Lemma 2.8. By Representation there is an \(s_\parallel \in {\sim }\{ s_\bot \}\) such that \(s \approx _{{\sim }\{ s_\bot \}} s_\parallel \). It follows that \(s_\parallel \in {\sim }{\sim }\{ s , s_\bot \}\).

It remains to show that \(s_\parallel \in P\). Suppose (towards a contradiction) that \(s_\parallel \not \in P\). Then \(s_\parallel \not \in {\sim }{\sim }P\). By Representation there is an \(s'_\parallel \in {\sim }P\) such that \(s_\parallel \approx _{{\sim }P} s'_\parallel \). One the one hand, since \(s'_\parallel \rightarrow s'_\parallel \), \(s_\parallel \rightarrow s'_\parallel \). On the other hand, since \(s_\parallel \not \rightarrow s_\bot \) and \(s_\bot \in {\sim }P\), \(s'_\parallel \not \rightarrow s_\bot \). Since \(s \approx _{{\sim }P} s_\bot \) and \(s'_\parallel \in {\sim }P\), \(s \not \rightarrow s'_\parallel \). Therefore, \(s'_\parallel \in {\sim }\{ s , s_\bot \}\). Since \(s_\parallel \in {\sim }{\sim }\{ s , s_\bot \}\), \(s_\parallel \not \rightarrow s'_\parallel \). As a result, we have got a contradiction. \(\square \)

Finally, we prove a lemma for building maximal opposite families used in the proof of Proposition 4.5.

Lemma A.7

Suppose that \(\mathfrak {F}\) satisfies Separation and Representation. Let I and J be two disjoint, non-empty sets, \(\{ P_i \mid i \in I \}\) a maximal opposite family in \(\mathfrak {F}\), \(i^* \in I\), \(\{ Q_j \mid j \in J \}\) a maximal opposite family in the subframe \(( P_{i^*} , {\rightarrow } )\) of \(\mathfrak {F}\) and \(K = ( I \setminus \{ i^* \} ) \cup J\). The following element in \(\wp (\wp (\Sigma ) \setminus \{ \emptyset \} )\) is a maximal opposite family in \(\mathfrak {F}\): for each \(k \in K\),

$$\begin{aligned} O_k = \left\{ \begin{array}{ll} P_k , &{} \text{ if } k \in I \setminus \{ i^* \} \\ Q_k , &{} \text{ if } k \in J \end{array} \right. \end{aligned}$$

Proof

The case when \(I = \{ i^* \}\) follows from Remark 3.2. It remains to consider the case when I is not a singleton.

In this case, it can be verified that \(\{ O_k \mid k \in K \}\) is an opposite family.

Before showing maximality, I claim that \({\sim }P_{i^*} = \bigcap \{ {\sim }Q_j \mid j \in J \}\): For one direction, for each \(j \in J\), \(Q_j \subseteq P_{i^*}\), so \({\sim }P_{i^*} \subseteq {\sim }Q_j\). Hence \({\sim }P_{i^*} \subseteq \bigcap \{ {\sim }Q_j \mid j \in J \}\). For the other direction, suppose (towards a contradiction) that \(\bigcap \{ {\sim }Q_j \mid j \in J \} \not \subseteq {\sim }P_{i^*}\). On the one hand, there is an \(s \in \bigcap \{ {\sim }Q_j \mid j \in J \}\) but \(s \not \in {\sim }P_{i^*}\). By Lemma A.6 there are \(s_\parallel \in P_{i^*}\) and \(s_\bot \in {\sim }P_{i^*}\) such that \(s_\parallel \in {\sim }{\sim }\{ s , s_\bot \}\). Since \(s \in \bigcap \{ {\sim }Q_j \mid j \in J \}\) and \(s_\bot \in {\sim }P_{i^*} \subseteq \bigcap \{ {\sim }Q_j \mid j \in J \}\), \(s_\parallel \in \bigcap \{ {\sim }Q_j \mid j \in J \}\). Thus \(s_\parallel \in P_{i^*} \cap \bigcap \{ {\sim }Q_j \mid j \in J \}\). On the other hand, since \(\{ Q_j \mid j \in J \}\) is a maximal opposite family in the subframe \(( P_{i^*} , {\rightarrow } )\) of \(\mathfrak {F}\), by Lemma A.3 and Corollary A.5\(\emptyset = \bigcap \{ {\sim }_{P_{i^*}} Q_j \mid j \in J \} = \bigcap \{ P_{i^*} \cap {\sim }Q_j \mid j \in J \} = P_{i^*} \cap \bigcap \{ {\sim }Q_j \mid j \in J \}\), which contradicts that \(s_\parallel \in P_{i^*} \cap \bigcap \{ {\sim }Q_j \mid j \in J \}\). As a result, \({\sim }P_{i^*} = \bigcap \{ {\sim }Q_j \mid j \in J \}\).

Now we prove maximality. By Lemma 3.4 it suffices to show that, for each \(k^* \in K\), \(O_{k^*} = \bigcap \big \{ {\sim }O_k \mid k \in K \setminus \{ k^* \} \big \}\). We consider 2 cases.

  • Case 1: \(k^* \in I \setminus \{ i^* \}\).

    $$\begin{aligned} O_{k^*} = P_{k^*}&= \bigcap \{ {\sim }P_i \mid i \in I \setminus \{ k^* \} \} \qquad \qquad \qquad (\text {Lemma } 3.4) \\&= {\sim }P_{i^*} \cap \bigcap \{ {\sim }P_i \mid i \in I \setminus \{ k^* , i^* \} \} \\&= \bigcap \{ {\sim }Q_j \mid j \in J \} \cap \bigcap \{ {\sim }P_i \mid i \in I \setminus \{ k^* , i^* \} \}\qquad \qquad \qquad \text {(the claim)} \\&= \bigcap \Big \{ {\sim }O_k \mid k \in K \setminus \{ k^* \} \Big \} \end{aligned}$$
  • Case 2: \(k^* \in J\).

    $$\begin{aligned} O_{k^*} = Q_{k^*}&= \bigcap \{ {\sim }_{P_{i^*}} Q_j \mid j \in J \setminus \{ k^* \} \} \qquad \qquad \qquad (\text {Lemma } 3.4) \\&= \bigcap \{ {\sim }Q_j \cap P_{i^*} \mid j \in J \setminus \{ k^* \} \} \qquad \qquad \qquad (\text {Lemma } A.3) \\&= P_{i^*} \cap \bigcap \{ {\sim }Q_j \mid j \in J \setminus \{ k^* \} \} \\&= \bigcap \{ {\sim }P_i \mid i \in I \setminus \{ i^* \} \} \cap \bigcap \{ {\sim }Q_j \mid j \in J \setminus \{ k^* \} \}\quad (\text {Lemma }3.4) \\&= \bigcap \Big \{ {\sim }O_k \mid k \in K \setminus \{ k^* \} \Big \} \end{aligned}$$

\(\square \)

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zhong, S. Quantum states: an analysis via the orthogonality relation. Synthese 199, 15015–15042 (2021). https://doi.org/10.1007/s11229-021-03453-5

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11229-021-03453-5

Keywords

Navigation