arXiv:0909.2359v3 [quant-ph] 15 Jun 2010 aepaudte“nepeain fteter o over for theory the that of paradoxes “interpretation” the myriad plagued have the and the function of wave collapse nonlocality, of al- entanglement, questions quantum system , Thus, this of complementarity. because uniqueness namely so fundamental , is the This displays ready time. one at freedom nItouto oCnitn unu Theory Quantum Consistent to Introduction An a eepandb tdigtesmls xml of - example single simplest a the system, studying mechanical quantum by con- and a confronted explained the be be must is mystery”, can presentation and “weirdness including this element its mechanics, quantum all original for of essence appealing only claimed the as The that tention be it author. find can present will that the they the with that does researchers hopes as and the in teachers theory, students, of munity its of words right!” the done In and “Copenhagen derived. is obscurity is precisely CQT the state what inventor to and of and assumed much is more!) as what no remove (but possible to as as mystery in way be formulated to a mechanics, meant quantum such is orthodox theory of version The a criticism. penetrat- and/or than analysis rather ing rather comments, of and object judgments the that cursory mostly within is Even CQT mechanics. however, community, be quantum specialists to of of circle appears foundations narrow it rather in a but of (1999), outside known Omnes little (Hartle, by also and (1993)(see 2008)) Hartle 2007, and elab- Gell-Mann further and by 1984) orated years (Griffiths, 25 Griffiths over Robert introduced by was ago theory The quan- non-relativistic mechanics. of tum clear- formulation the correct be most and to est CQT) (hereafter 2002) (Griffiths, Theory INTRODUCTION I. hsppri natmtt aiirz rae com- broader a familiarize to attempt an is paper This Quantum Consistent considers author present The sue oicroaeteeeeet na t in shown elements is these It incorporate to complementarity. assumed Bohr’s of “singl statement This elements ematical combining statements. sys and meaningless quantum framework mechanically excl single a quantum a of characterize to set confined fully consistent exhaustive be to must be an suffices provides can framework framework single a which A that for saying system “frameworks”. by theo a as The simultaneo of to the erties referred is years. , systems 25 of classical, th representations past to that opposed notion the as the over on quantum, based others Consiste mechanics, quantum and to (Copenhagen) Griffiths introduction elementary by an veloped presents paper This 2018) 24, October NY (Dated: York, New University, York New Physics, of Department Hohenberg C P nalgclmne.Teesnileeet fteter are theory the of elements spin- essential single The a system, manner. quantum mysterious logical other a and in collapse function wave , ASnmes 03.65.Ta numbers: PACS ½ ereo reo toetime. one at freedom of degree local ½ ereof degree etn,te itn nageet,macroscopic entanglements, distant then setting, rmwr ue stepeiemath- precise the is rule” framework e rmwr samxmlsto prop- of set maximal a is framework seitneo utpeincompatible multiple of existence us ydfie.Ti oini expressed is notion This defined. ly etrso unu eairfollow behavior quantum of features tQatmTer CT,a de- as (CQT), Theory Quantum nt a ftemcocpcdsrpinis description microscopic the if hat iw nta tdaswt el(..itisc properties intrinsic) (i.e. real with deals it that in view, olwn assumptions: following issue simplest fundamental the in- for the arises system. easily faced already relatively which has complementarity are of one that once add-ons corporated, become years 80 e.Aypeito ftetheory the of prediction Any tem. sv lentvs(SA,btno but (ESEA), alternatives usive nqeadmseiu etr of feature mysterious and unique e rmdffrn rmwrslasto leads frameworks different from rtepanduigtesimplest the using explained first sepandblw Q ae elsi on of point realistic a takes CQT below, explained As u omlto fqatmmcaisi ae nthe on based is mechanics quantum of formulation Our okrl” ob lcdtdi htfollows. what in frame- elucidated “single be the to of rule”, the work terms is in This expressed ity, incompatibil- or combined. complementarity or of principle compared can- be which not var- “frameworks”, in de- called described using scriptions, ways be re- incompatible can Instead alternative it ious that process. such or is hold: description system ality not exhaustive physical unique a does a of unicity not of is there principle The C. systems. physical about information probabilistic yields function wave quantum though the even to terministic so according evolve given is states This mechanical a multi- for at times. generalization ple property suitable specific with time, a study as under possess expressed system be will the should) that (and the can They probabilis- tic. intrinsically but are deterministic mechanics not quantum of predictions The B. appropriate properties an their in space. states and (Hilbert) by represented objects are Physical A. yi eso forthodox of version a is ry creigreuto,snethe since equation, Schroedinger de- 2 of physical systems, independent of observers or mea- The space of states possesses an inner product accord- surement. The essential feature of the theory is that its ing to which the vectors satisfy the orthonormality “quantum realism” posits multiple incompatible but co- conditions existing layers of reality, expressed in the theory as differ- + + − − ent “frameworks”. Any quantum mechanical statement w w = w w =1, (6a) h | i h | i must refer to one and only one framework (the “single w+ w− = w− w+ =0. (6b) framework rule”). This basic assumption is consistent h | i h | i with Bohr’s complementarity principle, but it is formu- The spin in direction w, say, is an represented lated without recourse to measurement or to the exis- by an Sw with the property tence of a classical regime and it is therefore more gen- + + − − Sw w = ½ w , Sw w = ½ w , (7) eral and can be more precisely formulated than Bohr’s | i | i | i − | i complementarity principle. which is expressed by saying that w+ and w− , are The above, of course, assumes that quantum mechan- | i | i eigenstates of Sw with eigenvalues ½ . The expecta- ics is an exact theory in the non-relativistic domain. ± tion value, or average value, Sw of Sw in an arbitrary Various authors have explored the possibility of physical h i corrections to , which would in par- state v is given by Sw = v Sw v , so it is ½ in the states| wi ± , respectively.h i Thish is| quite| i different± from the ticular bring about collapse of the when | i macroscopic measurements are made (see for example classical case described in Eqs.(1-2). Adler and Bassi (2009) and references therein), but we The basic assumption B stated in the Introduction is shall not consider this alternative in the present discus- embodied in the , which specifies that the prob- sion. ability that the operator Sw has the value ½ in the state v is given by | i + 2 2 Prob(Sw = ½)= w v = α , (8) II. A SPIN-½ PARTICLE |h | i| | | where Eq.(4) has been used.Correspondingly the proba- The classical representation of a spin of magnitude ½ bility of the value -½ is β 2 and by Eq.(5) those are the can be taken as a vector in 3-space with magnitude ½, only possibilities for S |,| since the probabilities add up i.e. a point on the surface of a sphere of radius ½. The w to unity. In accordance with assumption B, probability direction w, for example, leads to the spin is an intrinsic concept in the theory and not reducible to other notions. Sw = [Sxx + Syy + Szz], (1) Let us specify v = z and w = x or y. Then the commu- where tation relations between spin components Sx,Sy and Sz are realized by 2 2 2 Sx + Sy + Sz =1/4, (2) z+ = (1/√2)( x+ + x− )=(1/√2)( y+ y− ),(9a) x y z | −i | i | −i | i − | −i and , , are unit vectors forming a basis in 3-space. z = (1/√2)( x+ x )=(1/√2)( y+ + y ),(9b) describes the (orbit) of this | i | i − | i | i | i point on the surface of the sphere and a measurement, if If the system is in the state z+ , then the following is carried out appropriately, would verify the predictions of true: | i that theory. A feature of the states in Eq.(1) is that for S in the z-direction, for example, we have Prob(Sz = ½)=1, Prob(Sz = ½)=0, (10a) − Prob(Sx = ½)= ½, Prob(Sx = ½)= ½, (10b) Sz = ½, Sx =0, Sy =0, (3) − Prob(Sy = ½)= ½, Prob(Sy = ½)= ½. (10c) − i.e. the information regarding Sx,Sy and Sz is consistent 2 and mutually compatible. A general state can have any Note also that in contrast to Eq.(2), we have S =3/4 ½ h i value of Sx,Sy and Sz satisfying the constraint (2). for spin- in the quantum case. Quantum mechanically, the system is represented by states, also indexed by a direction w in 3-space, but these states exist in a discrete two-dimensional . III. THE ORTHODOX FORMULATION OF QUANTUM Any direction w defines a basis w+ , w− , such that any MECHANICS state of the system, corresponding| toi the| directioni v, say, According to the orthodox formulation of quantum me- may be written chanics (see, for example Landau and Lifshits (1958)), a v = v+ = α(v, w) w+ + β(v, w) w− , (4) system is “represented” by its wave function or state vec- | i | i | i | i tors which we denote by v , say. The physical meaning of | i with this state vector only becomes apparent when one makes a measurement, for example of Sw. Then the state of α 2 + β 2 =1. (5) the system “collapses” to one of the eigenstates w± of | | | | | i 3

Sw, with probability given by the Born rule, and the re- spin-½. The characteristics of a framework in this simple sult of the measurement is the corresponding eigenvalue. case result from those of the basis, namely (i) that any Thus, for example, if the state of the system is z+ and state can be represented and (ii) the disjoint possibilities | ± i I measure Sx, then the system goes to states x with associated with that framework (spin up or spin down) eigenvalues ½, respectively, each with probability| ½i. The have probabilities adding up to unity (certainty). Thus orthodox formulation± has been criticized by many au- a framework provides an Exhaustive Set of Exclusive Al- thors (see e.g. Bell (1990); Gell-Mann and Hartle (1993); ternatives (ESEA). It is exhaustive because of (i) and Griffiths (2002); Omnes (1999)) because the dynami- (ii), and exclusive because the basis vectors are mutually cal mechanism for the collapse remains undefined, and orthogonal. Note that although there are an infinite num- because of the heavy reliance on macroscopic measure- ber of different frameworks in which to represent a given ments and on the classical domain in defining the physi- state, since each one is exhaustive any quantum mechan- cal meaning of the theory. ical statement only makes sense (and thus can only be correct) relative to a particular framework. To illustrate the last statement let us associate with the state x+ + | i IV. COMPLEMENTARITY the “property” [x ] which for the present discussion we define to be the projection operator onto the state x+ . + − | i An important distinction between the classical and Then the properties [x ] and [x ] represent Boolean al- quantum formulations described above lies in the ques- ternatives and we can apply the ordinary rules of logic tion of compatibility between the values of Sx,Sy and Sz. to these operators. This logic refers to the x-framework. In the classical case the three statements in Eq.(3), for Alternatively, we can expand the state in question in the example, are all compatible, whereas the corresponding y-basis and employ the y-framework, but it is impossi- quantum mechanical statements in Eqs.(10) cannot be ble to define a framework that combines the properties + + combined into a single probabilistic formulation involving [x ] and [y ] in a consistent manner, since the projec- all three components of S. The reason is that each line tion operators do not commute. Indeed, any attempt + + of Eq.(10) refers to a different sample space and in con- to define the disjunction of properties [x ] [y ] (ei- ∨ + + ther Sx = ½ or Sy = ½) or the conjunction [x ] [y ] trast to classical mechanics these sample spaces cannot ∧ be consistently enlarged (“refined”), when the operators (Sx = ½ and Sy = ½) will fail, since there is no (x,y)- involved do not commute (see below). framework or (x,y) basis in the Hilbert space. In CQT The above is an aspect of complementarity, which is the essential feature of quantum mechanics, indeed the the essence of quantum mechanics and it is a good idea full mystery and weirdness of the theory, is captured by to incorporate such a fundamental feature in the very for- the existence of multiple frameworks, each one of which mulation of the theory, rather than as an after-thought provides an account of quantum mechanical “truth”, but once one has gotten into trouble. Complementarity is at different frameworks are mutually incompatible. A quan- the heart of the Heisenberg Uncertainty (or better, Inde- tum mechanical statement is only meaningful relative to terminacy) Principle, according to which two a particular framework and it is meaningless (quantum mechanical nonsense) relative to a different framework. represented by non-commuting operators (e.g. Sx and measured Thus asking “what is the probability that the system has Sy) not only cannot be simultaneously with ar- + bitrary precision, they cannot be defined simultaneously the property [x ]?” has no meaning in the y-framework. with arbitrary precision. Note that we shall often use the It is, however a perfectly meaningful question in the x- term “complementary” to refer to the incompatibility of framework. Indeed, for a system prepared in the state z+ I may make the (x-framework) statement Prob([x+]) any two noncommuting variables, not just to variables | i + such as x and p that are maximally incompatible. = ½ and the (y-framework) statement Prob([y ]) = ½, but as noted above, I cannot give meaning to the probability of “[x+] or [y+]”. Similarly, for a spin prepared in the state z+ , the statements in Eq. (10) “the system has V. CONSISTENT QUANTUM THEORY (CQT) the property| i [z+] with probability 1” and “the system has the property [x+] with probability ½” are not contra- How does one incorporate complementarity into the dictory but complementary, since they refer to different theory early on? A clue comes from the existence in the frameworks. This means that contrary to the classical Hilbert space of an infinity of bases, each one of which case there exists no overall framework of which each line can be used to represent any state of the system. For the in (10) would be a component. simple case of a spin-½ system we define a “framework” as any basis in which we choose to represent a given state In concluding this section let us note that although of the system. [This definition will be generalized in Sec. each framework is associated with one and only one ba- VIII below]. Since according to Eq.(4) an arbitrary quan- sis in the Hilbert space, a framework and a basis are tum state v can be expanded in an infinite number not identical. Indeed, since the full wave function can of ways by| choosingi an appropriate basis w± , we see be expanded in any basis the different bases are in some that the different frameworks are merely different| i ways sense equivalent. A framework, on the other hand, iden- of representing an arbitrary state in the Hilbert space of tifies the properties (eigenstates) associated with the cor- 4 responding basis, but only retains the real probability for EPR2: a criterion of separability (or local- those properties and discards the remaining information. ity), which was articulated more clearly in Thus different frameworks capture different real aspects subsequent presentations of the argument by of a quantum system and a full description of that system Einstein (Born and Einstein, 1971; Einstein, requires the set of all frameworks. 1954, 1969), according to which “the real fac- tual situation of system B is independent of S what is done with system A if the latter is VI. MEASUREMENTS spatially separated from theS former”.

As is well known any component of spin can be With these two assumptions EPR showed that the quan- measured using a suitable Stern-Gerlach apparatus (see tum mechanical wave function does not provide a com- Griffiths (2002) or Landau and Lifshits (1958)). Notice, plete description of physical reality. however, that the preceding discussion of CQT made no CQT modifies EPR1 by replacing this “classical real- reference to measurements. This is because from the ism” assumption by “quantum realism”, whereby phys- point of view of CQT measurements play no special role, ical properties of a system are real relative to a well- just as in classical mechanics. This means that since mea- defined framework, but properties belonging to incom- surements are just one type of physical interaction they patible frameworks are not simultaneously real. As rec- must also be understood and interpreted within an ap- ognized by the authors themselves, the EPR conclusion propriate framework. Although a proper description of then does not follow. To quote them “... one would not such physical interactions requires the more general for- arrive at our conclusion if one insisted that two or more malism of Sec.VIII below, since the system must now in- physical quantities can be regarded as simultaneous el- clude a measurement apparatus, we can anticipate that a ements of reality only when they can be simultaneously measured or predicted” (italics in the original). Whether measurement of Sx, for example, needs to be interpreted in the x-framework. It will not be informative if analyzed one follows EPR or CQT in defining “physical reality”, the issue in contention concerns the simultaneous reality in the z-framework. A As noted above, a measurement of S is a physical of quantities pertaining to the same subsystem (i.e. Sx x A A process which causes its different eigenstates to be corre- and Sz , both of which refer to the spin S ), as opposed lated with states of a macroscopic measurement appara- to correlations or “entanglements” between quantities re- A B tus. These correlations are not problematic when viewed ferring to distant spins S and S . Note also that Bohr’s in the appropriate framework. Rather than to say response to EPR (Bohr, 1935, 1937; Einstein, 1969) in- volved objections to EPR1, phrased in terms of incom- “given that the system has been prepared in patible measurements, and not any denial of EPR2. the state z+ , I now make a measurement of In a sense the main criticism leveled by Einstein, in | i + + Sx, which collapses the state z to x or EPR and elsewhere, against orthodox quantum mechan- − | i | i x , each with probability ½”, ics concerned the notion of collapse of the wave function | i CQT says brought about by a measurement. In the orthodox for- mulation such a collapse occurs even if the physical action “in the x-framework any state has two dis- is carried out arbitrarily far from the system being mea- joint possibilities or properties [x+] and [x−], sured. It is the effects associated with collapse, e.g. that + A so for a system prepared in the state z a Sz comes into existence or ceases to exist through a mea- | i B measurement of Sx, when viewed in the x- surement carried out on the arbitrarily distant spin S , framework, will reveal one or the other of that Einstein termed “spooky action at a distance”, not these properties, each with probability ½. In the correlations that allow one to determine the value A B the z-framework, on the other hand, I am un- of Sz by making a measurement of Sz . Such distant able to interpret an Sx measurement”. correlations (within a given framework) occur classically as well as quantum mechanically. According to CQT it These issues will be taken up again in Sec. IX, below, is only the mixing of correlations in different quantum once the more general formulation of CQT has been pre- frameworks that produces puzzling or incorrect results. sented. Classical entanglement was described by John Bell us- ing the example of his colleague Dr. Bertlmann (see p. 139 of Bell (2004)), who always wore socks of different VII. EINSTEIN, BOHR AND BELL colors. Thus if I see a black sock on Dr. Bertlmann’s The Einstein, Podolsky, Rosen (EPR) (Einstein et al., right foot I know for sure that there is a white sock on 1935) argument starts from two basic assumptions: his left foot, without having to look. Such entanglements also operate between classical particles, in that I can de- EPR1: a (classical) reality criterion accord- termine the spin of particle A by measuring the spin of ing to which a physical quantity is real if its particle B, no matter how far away, provided the two value can be predicted with certainty without particles were in a state of zero angular at in any way disturbing the system; and some earlier time and they drifted away without further 5 interactions. An example of distant classical correlations inequality [see e.g. Ballentine and Jarrett (1987) or more (although not classical entanglements) is the statement recently Griffiths (2009) and Tresser (2010)]. In particu- that if my daughter has a baby in Sydney, Australia, I lar the existence of λ variables in Eq.(11) that fully spec- instantaneously become a grandfather in NY! ify the state of a physical system depends essentially on Turning to quantum mechanics, within a given frame- some assumption additional to locality, which has been work entanglement also operates classically. What is spe- variously named “predictive completeness”, “classical (or cial about two quantum mechanically entangled parti- weak) realism”, or “counterfactual definiteness”, as well cles is that the correlation is effective in each and ev- as other designations. According to these authors (and ery framework, but different correlations that cannot be to CQT) it is this additional assumption that is contra- incorporated into a single framework are not simultane- dicted by quantum mechanics, and referring to Eq.(11) ously meaningful. simply as a locality assumption is at best misleading. As is well known it was Bell (1964) who showed that Indeed, the violation of Eq. (11) by quantum mechan- the two EPR assumptions not only lead to a puzzling re- ics is due to the incompatibility between different local A A sult, but in fact they contradict the quantitative statis- choices for a, i.e. Sx or Sz (or different choices for b), tical predictions of quantum mechanics. Indeed, starting rather than any “spooky interaction” between the distant from the two EPR assumptions, Bell was able to derive spins SA and SB, so it is unrelated to the locality prop- what is known as the Bell inequality, which is violated erty EPR2. In particular, Griffiths (2009) argues that by quantum correlations. If one believes quantum me- there are no quantum variables λ that satisfy Eq.(11) chanics, one is thus forced to reject either classical real- if the different choices for a (or for b) do not commute. ism, or locality, or both. A number of authors, includ- What is at issue here is not the fact that Eq.(11) and ing Bell himself (Bell, 1964, 2004), Bohm (1952), many the Bell inequality that follows from it are violated by philosophers and commentators on quantum mechanics quantum mechanics. The issue is to understand the ori- (Albert, 1994; Greene, 2004; Maudlin, 2002) as well as gin of the violation and whether it is associated with a current researchers (see e.g. Ekert (1991); Gisin (2009)), failure of realism or completenes (EPR1), or of separa- favor rejecting locality, but CQT most definitely opts for bility (EPR2). Needless to say, there is no consistent retaining locality (EPR2) and rejecting classical realism framework for which Eq.(11) is satisfied. (EPR1), since the latter violates the single-framework The present author believes that had Einstein been rule. aware of Bell’s result, which follows from EPR1 and In our view the difference of opinion arises in part be- EPR2, he too would have retained locality, since it is cause of a confusion regarding the assumptions needed so closely tied to the requirements of relativity, and he to prove the Bell inequality. Although Bell (1964) ini- would have had to reject classical realism. It is per- tially presented the argument as simply drawing a fur- haps permitted to fantasize that Einstein would have ther conclusion from the two EPR assumptions, in later embraced the quantum realism of CQT, as the best way presentations (see Bell (2004), p.232) he based the argu- to reconcile quantum mechanics (whose predictions Ein- ment entirely on a condition he called “local causality”, stein did not doubt) with the requirements of locality. often referred to as “Bell locality” (Norsen, 2006). Its According to CQT reality is relative to a framework just mathematical expression is as simultaneity is relative to a reference frame in special relativity. Prob(A B,a,b,λ) = Prob(A a, λ), (11a) | | Prob(B A,a,b,λ) = Prob(B b, λ), (11b) | | Prob(A, B a,b,λ) = Prob(A B,a,b,λ)Prob(B a,b,λ) VIII. | | | = Prob(A a, λ)Prob(B b, λ), (11c) | | In order to provide a complete formulation of quan- where Prob(C D) = Prob(C,D)/Prob(D) is the condi- tum mechanics the foregoing notions must be general- tional probability| of C given D and Prob(C,D) is the ized to consider systems with arbitrary numbers of de- joint probability of C and D. In Eqs.(11a-c) A and B grees of freedom as well as their time dependence. This are measurement outcomes for measurements of types a generalization is the CQT formulated by Griffiths (1984) and b, respectively, on systems A and B, and λ de- and elaborated by others (Gell-Mann and Hartle, 1993; notes a set of additional variablesS that mayS be required Omnes, 1999), as described in detail in the book by to specify fully the state of the joint systems. Although it Griffiths (2002). The general theory is still based on is not completely clear from the discussion in Bell (2004) the fundamental principle that what characterizes any referred to above what precise assumptions went into the quantum system is the existence of multiple incompatible justification of Eq.(11), Bell and his followers have con- frameworks, which are a generalization of the frameworks cluded from the violation of the Bell inequality that fol- defined above. lows from (11) that it is the locality assumption that Let us first define a “history” as a set of “events” or is violated by quantum mechanics. On the other hand “properties”, represented by projection operators on the many authors have over the years pointed out that local- Hilbert space at a sequence of times t1, ...tn. We then ity alone does not justify Eq. (11) or therefore the Bell consider a collection, or “family” of histories as a candi- 6 date for a (consistent) framework. The probability of a where given history is calculated using a generalization of the α = Pˆ(αn)...Pˆ(α ) ψ , (15) Born rule, as well as the dynamical properties of the sys- | i 1 | 0i tem, whose states evolve according to the Schroedinger ˆ equation governed by the full Hamiltonian of the system and P (αk) is a projection operator at time tk in the (see below). In contrast to the spin-½ case at a single Heisenberg representation. The probability of the his- time, where frameworks were more or less automatically tory C(α) is then given by a generalization of the Born consistent, it is in general necessary to impose stringent rule consistency conditions in order for a given family of his- Prob[C(α)] = C(α), C(α) = α α . (16) tories to provide valid quantum predictions. The condi- h i h | i tions for a family to constitute a consistent framework Let us now consider a set or family of histories are: ( α )= C(α(1)), C(α(2)), ...C(α(m)) , (17) (i) The sum of probabilities of the histories in F { } { } the family is unity; where each of the C(α(k)) is a history, i.e. a product of (ii)Two distinct histories are mutually or- n projection operators as in Eq.(12). In order to be able thogonal, i.e. their product vanishes. to assign probabilities in a consistent manner, the family ( α ) must satisfy the following conditions: Each framework can again be described as an Exhaus- F { } m tive Set of Exclusive Alternatives (ESEA), exhaustive (k) because of (i) and exclusive because of (ii), with a suit- X C(α )= Iij , (18) able definition of the product. Also, just as for spin-½, k=1 frameworks cannot be combined, i.e. you obtain incor- where Iij is the n n unit in the history Hilbert rect predictions if you take a subset of histories from one space and × framework (with probabilities adding up to p, say) and ′ combine them with another subset of histories from an- C(α(k)), C(α(k )) =0, for k = k′, (19) other framework (with probabilities 1 p), in violation of h i 6 the single-framework rule. We shall use− the terms family which expresses the orthogonality condition (ii). A fam- and framework somewhat interchangeably, but in general ily of histories satisfying Eqs.(18) and (19) is called a reserve the latter term for families whose histories satisfy “consistent framework” (or framework for short). Equa- the consistency conditions. Another term for frameworks tion(18) ensures that the framework is exhaustive and is “realms” (Gell-Mann and Hartle, 1993). Eq.(19) implies that the different histories are exclusive, Specifically, let t0,t1, ..., tn be a discrete set of times i.e. they do not “interfere”, thus permitting an interpre- and let a “property” or “event” αj at time tj be defined tation in terms of standard (Boolean) probabilities within as a state or set of states in the Hilbert space of the each framework. system under study, or as the projection operator P (αj ) The hallmark of a quantum, as opposed to a classical, onto that state or set of states, for which we shall use the system is the existence of more than one (in fact an in- same notation. A “history” is then defined as a sequence finite number of) framework(s). Any statement about a of projection operators whose tensor product is quantum system is expressed in the form

(k) C(α) = C(α0, α1, ..., αn) Prob[C(α )] = pk, (20)

= P (α ) P (α ) ... P (αn), (12) th 0 ⊙ 1 ⊙ ⊙ i.e. the probability of the k history in the framework ( α ) is p . Such statements refer to “real” histories of where the notation is designed to indicate that each k theF { form} (12), which consist of sequences of “real” prop- P (α ) acts in a copy⊙ of the system Hilbert space at time k erties, represented by projection operators onto states t , so that C(α) is an operator in the “history Hilbert k α of the system. The reality of these properties, how- space” consisting of n replicas of the original Hilbert j |ever,i is relative to a particular framework. space. In general the product of two histories must be For illustration let us first consider once more the free defined as an operation in this history Hilbert space (see spin-½ discussed above. We denote the projector onto the Ch.11 of Griffiths (2002)). If the system under study is + state w at time tj by assumed to be in the pure state ψ0 at time t0, however, | i | i i.e. if the projector P (α ) is given by + + + + 0 P ( w ,tj ) = [w ]= w (tj ) w (tj ) . (21) | i j | ih | P (α )= ψ ψ [ψ ], (13) 0 | 0ih 0|≡ 0 A history with times t0,t1, ...tn will be written as then it is possible to define the product of two histories C = [u0] [v1] ... [wn], (22) in terms of “chain kets” in the original Hilbert space as ⊙ ⊙ ⊙ in the history Hilbert space, where [ui], [vj ], [wk] denote C(α), C(β) = α β , (14) projection operators onto states associated with various h i h | i 7 directions in 3-space at times ti,tj ,tk, respectively, e.g. whose probability is unity, where [ψj ] is the projector + [ui]= P ( u ,ti), using the same notation as in Eq.(13) onto the state ψ advanced by the unitary Schroedinger | i | 0i above. time dependence to the time tj . We refer to the family It can be shown that any exhaustive family of two- (26) as the “unitary framework” (see Sec. IX below). time histories automatically satisfies the consistency con- Another consistent framework splits the histories at t = ditions (18) and (19). For three-time histories, on the t1 and then continues unitarily to t = t2 other hand, the consistency conditions are not necessar- (1) + − + − ily satisfied. An example is the family consisting of the C = [ψ0] [zA1][zB1] [zA2][zB2], (27a) ⊙ − ⊙ − histories C(2) = [ψ ] [z ][z+ ] [z ][z+ ], (27b) 0 ⊙ A1 B1 ⊙ A2 B2 (1) + + + C = [z0 ] [x1 ] [z2 ], (23a) to yield two histories each with probability ½. Alterna- ⊙ ⊙ − C(2) = [z+] [x+] [z ], (23b) tively, the split could occur at t = t2, and by spherical 0 ⊙ 1 ⊙ 2 symmetry of ψ the direction z in Eqs.(27) can be re- (3) + − + | 0i C = [z0 ] [x1 ] [z2 ], (23c) placed by any other direction w. As an example of an ⊙ − ⊙ − C(4) = [z+] [x ] [z ]. (23d) inconsistent family we show the set of 16 histories 0 ⊙ 1 ⊙ 2 + − + − The framework consisting of the histories in (23) can be [ψ0] [xA1], [xA1] [zB1], [zB1] (1) (3) ⊙ { + − }{ + − } shown to be inconsistent since the histories C and C [zA2], [zA2] [xB2], [xB2] , (28) [or C(2) and C(4)] are not mutually orthogonal, i.e. they ⊙ { }{ } interfere. It is the presence of the final [z+] operator in where the notation C,D E, F denotes the four prod- 2 { }{ } Eqs.(23a) and (23c) that brings about the nonorthogo- ucts CE,DE,CF,DF . It can then be verified that the nality or interference between the histories C(1) and C(3). family (28) is inconsistent since there are four histories + − − If [z2 ] is replaced by a unit operator, on the other hand, ending in [zA2][xB2] and these are not mutually orthog- then the histories C(1) and C(3) become orthogonal. An- onal. other way to make the framework (23) consistent is to Finally, it should be noted for clarity that the histories apply an appropriate magnetic field during the time in- of CQT are different from the histories in Feynman’s path (2) terval t0

C = [ψ ] [ψ ] [ψ ], (26) [ψ ] [ψ ] ... [ψn], (29) 0 ⊙ 1 ⊙ 2 0 ⊙ 1 ⊙ ⊙ 8 whose probability is unity. This is a valid prediction system in the state z+ is to say “I will accept a super- of quantum mechanics, but unless the final state is the position of x+ and| x−i for microscopic states, since we eigenstate of some straightforward observable it is also know from| experiencei | thati the microworld is weird, but not a particularly useful prediction, since it contains no for macroscopic (cat) states, that is impossible”. The an- accessible physical information. In particular the values swer from CQT is that in neither case (micro or macro) of any observables are in general undefined in this frame- is the “superposition” a valid quantum mechanical de- work. scription. In the z-framework neither x+ nor x− is | i | i There is, however, a framework that is close to the defined (no cats) and in the x-framework we have either unitary one, which we can call the “collapse framework”, Sx = ½ (live cat) or Sx = ½ (dead cat), each with h i h i − which is the one employed in the orthodox formulation probability ½. For the macroscopic case the analogue of of quantum mechanics. It consists of the histories enu- the z-framework is the unitary framework and it has no merated by the index k, simple physical interpretation, and certainly no cats. The analogue of the x-framework is the set of (Boolean) al- A Ck = [ψ0] [ψ1] ... [ψn−1] [Akn], (30) ternatives -live cat or dead cat -, with appropriate prob- ⊙ ⊙ ⊙ ⊙ abilities for each. th where [Akn] is the projector onto the k eigenfunction of the operator A at time tn, with eigenvalue ak and A is some observable. Since the eigenfunctions of A form XI. DECOHERENCE AND THE A an orthonormal basis, the family Ck is an appropriate consistent framework. If the Hamiltonian of the system The effects of a random environment on physical sys- is such that a measurement of A is carried out at the tems and measurement instruments, termed decoherence time tn, then this framework embodies the orthodox no- (Zeh, 1970; Zurek, 2003), are often cited as explaining tion that in all histories the system is described by ψ(t) | i how one passes from the microscopic to the macroscopic from t = t0 to t = tn−1, and that the system then “di- and thus how the collapse invoked in the orthodox formu- vides” into the different eigenfunctions of A when a mea- lation of quantum mechanics is to be interpreted. Deco- surement of A is made at t = tn. In CQT this collapse herence is also invoked to understand the “measurement framework is one valid representation of the system, but problem” and the classical limit of quantum mechanics. by no means the only one and it does not signify that From this point of view there is nothing special about any physical collapse has taken place. measurements: they are a particular type of physical in- teraction that needs to be explained by the theory, rather than being an irreducible part of its formulation as in the X. QUANTUM PARADOXES orthodox or Copenhagen viewpoints. It has been pointed out, however, that although deco- CQT, as the name indicates, is designed to provide a herence does make plausible the passage from quantum consistent account of the predictions of quantum mechan- to classical domains, when it is applied to the full quan- ics, grounded in a realistic view of the microscopic world, tum mechanical wave function as a representation of the that recognizes the profoundly counter-intuitive nature of state of the system, it still does not by itself remove the complementarity but defines its consequences precisely. or the paradox of Schroedinger’s The theory thus contains no paradoxes, so it must in Cat (Adler, 2003; Leggett, 2002). some way “explain” the well-known quantum paradoxes CQT agrees with the above statements since, as stated that have been the subject of so much discussion in the previously, it attaches no special significance to macro- literature. Generally speaking, each paradox can be seen, scopic measurements. Moreover, the theory recognizes from the point of view of CQT, as a more or less subtle that decoherence applied in the unitary framework does violation of the single-framework rule, as explained in indeed not resolve the paradox of Schroedinger’s Cat detail in Chapters 20-25 of the book by Griffiths (2002). or the measurement problem. The relevant question, We have already indicated how this occurs for the EPR however, is “what is the effect of decoherence not on and Bell “paradoxes”, and we shall confine our further wavefunctions but on histories, in the different frame- remarks to a brief discussion of the Schroedinger Cat works?” Although the answer has not been fully worked paradox, since it has been cited by Leggett (2002, 2005) out, the problem has been extensively studied, primarily as the most important problem with orthodox quantum by Gell-Mann and Hartle (1993), and by Omnes (1999) mechanics. and the overall picture is the following (see also Ch 26 of The Schroedinger Cat paradox can be thought to arise Griffiths (2002)). because one is implicitly thinking in terms of the unitary Decoherence in general destroys interference effects framework, and at the same time thinking of the system and it can thus render certain families of histories (frame- as containing a superposition of live and dead cats. Ac- works) consistent, when in the absence of decoherence cording to CQT, however, there are no cats at all in the they would be inconsistent. It can then be argued, and unitary framework and in the cat-framework there are shown in detail for simple models, that a large class of no superpositions. The simple analogue for the spin-½ ensuing coarse-grained histories (as opposed to wavefunc- 9 tions), are close to the corresponding classical trajectories consistent framework. All of the difficulties encountered for the system and that the corresponding probabilities by previous formulations and interpretations (the para- are close to unity, thus justifying the of clas- doxes and the weirdness) are encapsulated (not elimi- sical mechanics as emerging from quantum mechanics. nated!) in the notion that quantum systems necessar- This does not mean that all frameworks are transformed ily possess multiple incompatible frameworks. Once this into sets of quasiclassical orbits, for example the unitary notion has been accepted (“digested” might be a better framework certainly is not, but the latter also does not in term) the theory unfolds in a logical manner without fur- general provide a useful set of predictions. The general ther paradoxes or mysteries. In CQT the wavefunction situation appears to be that in the classical limit, frame- is by no means unimportant, since as mentioned above works are either inconsistent (thus meaningless), uninfor- it encodes all the dynamical information about a system mative (e.g. the unitary framework), or they approach and it is thus essential for calculating any meaningful classical behavior. properties, but it does not provide a direct universal rep- In this way CQT naturally leads to the emergence of resentation of . It is true, of course, that quasiclassical frameworks on the coarse-grained scale of in the unitary framework described in Sec. IX the wave macroscopic phenomena, coexisting with a multiplicity function constitutes a valid history, but that history is in of incompatible but individually consistent frameworks general not useful for answering physical questions about on the microscopic scale. As noted, such a division be- the system. tween the classical and quantum domains was a starting In a sense, CQT can be considered to vindicate EPR in assumption of the orthodox and Copenhagen formula- that it confirms the statement that the wavefunction does tions. Note that decoherent histories are a special, albeit not provide a complete representation (“description”) of important, example of consistent histories, but the phys- physical reality. CQT also follows EPR in seeking a real- ical phenomenon of decoherence is not necessary in order istic formulation of quantum mechanics, but it replaces to formulate the theory, for example it is not involved in the classical realism of EPR1 by the quantum realism of our description of the single spin-½ system in Sec. II, so multiple coexisting but incompatible frameworks. In that Griffiths’s consistency conditions represent a more gen- sense CQT reconciles Einstein and Bohr, but without the eral formulation of the approach. latter’s resort to measurement and classical apparatus in the definition of quantum reality. It may also be argued that CQT satisfies the require- XII. CONCLUDING REMARKS ments put forth by Bell (1990) in his criticism of the or- thodox and Copenhagen viewpoints. CQT is not based It appears that the main philosophical difference be- on measurement, it does not require an artificial sepa- tween CQT and most other formulations of quantum ration (a “cut”) between the classical and quantum do- mechanics lies in the role ascribed to the wavefunction. mains and it needs no “interpretation”, since the theory Since the wavefunction encodes all the information about is defined by the equations themselves. a system it is natural to interpret it to “represent” the We have already discussed the sense in which the vi- system, just as the phase space point and its orbit rep- olation of the Bell inequality, which results from the resent a classical system. It was recognized early on by violation of Eq.(11), should be interpreted as negating Bohr and his colleagues, however, that the wavefunction classical realism rather than locality. As emphasized by could not provide a realistic representation of a quantum Griffiths (2009), quantum mechanics certainly has non- system, isolated from , i.e. from some mea- local states and frameworks, but then so does classical surement apparatus. This idea was made precise by EPR, mechanics, for example a plane wave state is surely non- who concluded that the wavefunction does not provide local. What does not occur in quantum mechanics, any a “complete description” of a quantum system. Bohm more than in classical mechanics, is true nonlocality in (1952) and Bell (1964) then showed that if one insists the sense of nonlocal influence of one localized subsystem on a (classically) realistic representation, it follows that on another remote localized subsystem. This is why the quantum systems will inevitably display nonlocal influ- pervasive use of the term “nonlocal” to describe quan- ences that violate the spirit, if not the letter, of relativity. tum mechanics and particularly the violation of Eq.(11), In its most extreme form the insistence that the wave seems to us to be maximally misleading. function provide a complete representation of quantum Let us comment briefly on the concept of “entangle- reality leads to the “relative state” or “many worlds” ment”, which is cited by many authors as the signature theory of Everett (1957). The dizzying complexity of property of quantum systems that illustrates its funda- even the simplest systems in this theory seems to the mental nonlocal character. From the point of view of present author to provide ample motivation for seeking CQT, entanglement is a description of how properties a different formulation. contained in framework appear in the language of CQT, on the other hand, retains the requirement of framework . The phenomenonF is particularly striking realism, but it abandons the notion that it is the wave when propertiesG referring to two widely separated sub- function that “represents” the system under study, sub- systems factorize in framework , while those same two stituting instead histories, provided these are part of a subsystems contribute to a singleF property in framework 10

. An example is the EPR pair considered in Eq.(24) of the Aspen Center for Physics where this paper was above,G for which one can calculate the probability that largely written. spins A and B have a relative angle θ using the framework (25), whereas these properties are fully entangled and thus undefined in the unitary framework based on (24). References Similarly, in a two-slit experiment the framework dis- playing the interference pattern does not contain states Adler, S. L., 2003, Studies in History and Philosophy of Mod- 34 in which the particle goes through either slit, since these ern Physics , 135. 325 are entangled in that framework. Strictly speaking, CQT Adler, S. L., and A. Bassi, 2009, Science , 275. Albert, D., 1994, Quantum Mechanics and Experience (Har- does away with entanglement and replaces it with incom- vard University Press). patible frameworks, i.e. the knowledge that even though Ballentine, L. E., and J. P. Jarrett, 1987, American Journal a framework is exhaustive it is not “complete”, since it of Physics 55, 696. allows for the existence of other frameworks. Bell, J. S., 1964, Physics 1, 195. Quantum reality can be succinctly defined as “any Bell, J. S., 1990, in Sixty-Two Years of Uncertainty, edited property or set of properties of a system for which quan- by A. I. Miller (Plenum Press, NY). tum mechanics can calculate a probability”. Note that Bell, J. S., 2004, Speakable and Unspeakable in Quantum Me- quantum mechanics does not answer the question “which chanics (Cambridge University Press). 85 real property or event actually occurs?” That question Bohm, D., 1952, Phys. Rev. , 166. Bohr, N., 1935, Phys. Rev. 48, 696. can be considered “excess baggage” (Hartle, 2007). A 4 further essential insight of CQT is that probabilities can Bohr, N., 1937, , 289. Born, M., and A. Einstein, 1971, The Born-Einstein Letters only be defined if a suitable sample space has been spec- (Walker and Co., NY). ified, and quantum systems are characterized by a mul- Einstein, A., 1954, Ideas and Opinions (Crown Publishers, tiplicity of overlapping and incompatible sample spaces, NY). called frameworks. Any statement about quantum real- Einstein, A., 1969, in :Philosopher-Scientist, ity must specify the framework (or set of frameworks) to edited by P. Schilpp (The Open Court Publishing Co. La which it applies. Salle, IL, 3rd Edition). We conclude by quoting Einstein, A., B. Podolsky, and N. Rosen, 1935, Phys. Rev. (Feynman et al., 2006): 47, 777. Ekert, A., 1991, Phys. Rev. Lett. 67, 661. In this chapter we shall tackle immediately Everett, H. I., 1957, Rev. Mod. Phys 29, 454. the basic element of the mysterious behav- Feynman, R. P., and A. R. Hibbs, 1965, Quantum Mechanics ior in its most strange form. We choose to and Path Integrals (McGraw-Hill, NY). The examine a phenomenon [two-slit interference] Feynman, R. P., R. B. Leighton, and M. Sands, 2006, Feynman Lectures on Physics (Addison-Wesley, Reading, absolutely which is impossible, impossible, to MA). explain in any classical way, and which has in Gell-Mann, M., and J. B. Hartle, 1993, Phys. Rev. D47, 3345. it the heart of quantum mechanics. In real- Gisin, N., 2009, Science 326, 1357. ity, it contains the only mystery. We cannot Greene, B., 2004, The Fabric of the Cosmos (A. Knopf, NY). make the mystery go away by “explaining” Griffiths, R. B., 1984, J. Stat. Phys. 36, 219. how it works. We will just tell you how it Griffiths, R. B., 2002, Consistent Quantum Theory (Cam- works. In telling you how it works we will bridge University Press). have told you about the basic peculiarities of Griffiths, R. B., 2009, eprint quant-ph/0908-2914. all quantum mechanics. Hartle, J. B., 2007, J. Phys. A Math. and Theor. eprint quant- ph/0610131v3. Feynman’s point of view is very much in line with that Hartle, J. B., 2008, eprint quant-ph/0806.3776v3. of CQT, in particular the statement that the only mys- Landau, L. D., and E. M. Lifshits, 1958, Quantum Mechanics, tery of quantum mechanics can be fully displayed using Nonrelativistic Theory (Addison-Wesley, Reading, Mass). Leggett, A. J., 2002, J. Phys. C. Cond. Mat. 14, R425. the example of a single electron in a two-slit interfer- Leggett, A. J., 2005, Science 307, 871. ence experiment, which displays the wave-particle du- Maudlin, T., 2002, Quantum Non-locality and Relativity ality. Note in particular the absence of any reference (Blackwell, Maulden Mass). to entanglement of distant particles or to nonlocality of Norsen, T., 2006, Foundations of Physics Letters 19, 633. quantum behavior. Omnes, R., 1999, Understanding Quantum Mechanics (Princeton University Press). Tresser, C., 2010, European Journal of Physics (to appear). Acknowledgments eprint quant-ph/0608008v2. Zeh, H. D., 1970, Found. Phys. 1, 69. 75 The author wishes to thank Robert Griffiths, James Zurek, W. H., 2003, Rev. Mod. Phys. , 715. Hartle, Tony Leggett, David Mermin and especially Richard Friedberg for useful discussions and comments on the manuscript. He also acknowledges the hospitality