Connecting real and hyperarithmetical analysis
Abstract.
Going back to Kreisel in the Sixties, hyperarithmetical analysis is a cluster of logical systems just beyond arithmetical comprehension. Only recently natural examples of theorems from the mathematical mainstream were identified that fit this category. In this paper, we provide many examples of theorems of real analysis that sit within the range of hyperarithmetical analysis, namely between the higher-order version of - and weak--, working in Kohlenbach’s higher-order framework. Our example theorems are based on the Jordan decomposition theorem, unordered sums, metric spaces, and semi-continuous functions. Along the way, we identify a couple of new systems of hyperarithmetical analysis.
Key words and phrases:
Higher-order arithmetic, hyperarithmetical analysis2010 Mathematics Subject Classification:
03B30, 03F351. Introduction
1.1. Motivation and overview
The aim of this paper is to exhibit many natural examples of theorems from real analysis that exist in the range of hyperarithmetical analysis. The exact meaning of ‘hyperarithmetical analysis’ and the previous boldface text is discussed in Section 1.3, but intuitively speaking the latter amounts to being sandwiched between known systems of hyperarithmetical analysis or their higher-order extensions. We shall work in Kohlenbach’s framework from [kohlenbach2], with which we assume basic familiarity.
We introduce some necessary definitions and axioms in Section 1.2. We shall establish that the following inhabit the range of hyperarithmetical analysis.
-
•
Basic properties of (Lipschitz) continuous functions on compact metric spaces without second-order representation/separability conditions, including the generalised intermediate value theorem (Section 2).
-
•
Properties of functions of bounded variation, including the Jordan decomposition theorem, where the total variation is given (Section 3).
-
•
Properties of semi-continuous functions and closed sets (Section 4.1).
-
•
Convergence properties of unordered sums (Section 4.2).
These results still go through if we restrict to arithmetically defined objects by Theorem 2.8. To pinpoint the exact location of the aforementioned principles, we introduce a new ‘finite choice’ principle based on finite--AC0 from [gohzeg] (see Section 1.2), using Borel’s notion of height function ([opborrelen4, opborrelen5]).
Finally, as to conceptual motivation, the historical examples of systems of hyperarithmetical analysis are rather logical in nature and natural examples from the mathematical mainstream are a relatively recent discovery, as discussed in Section 1.3. Our motivation is to show that third-order arithmetic exhibits many robust examples of theorems in the range of hyperarithmetical analysis, similar perhaps to how so-called splittings and disjunctions are much more plentiful in third-order arithmetic, as explored in [samsplit]. In this paper, we merely develop certain examples and indicate the many possible variations.
1.2. Preliminaries
We introduce some basic definitions and axioms necessary for this paper. We note that subsets of are given by their characteristic functions as in Definition 1.2, well-known from measure and probability theory. We shall generally work over -defined right below- as some definitions make little sense over the base theory . We refer to [kohlenbach2] for the latter.
First of all, full second-order arithmetic is the ‘upper limit’ of second-order RM. The systems and are conservative extensions of by [hunterphd]*Cor. 2.6. The system is plus Kleene’s quantifier (see e.g. [dagsamXIV] or [hunterphd]), while is plus for every ; the latter axiom states the existence of a functional deciding -formulas in Kleene normal form. The system is a -conservative extension of ([yamayamaharehare]), where is also called the Suslin functional. We also write for where the latter is as follows
() |
The system is a conservative extension of by [hunterphd]*Theorem 2.5. Over , is equivalent to , where the latter expresses the existence of Feferman’s (see [kohlenbach2]*Prop. 3.9), defined as follows for all :
The following schema is essential to our enterprise, as discussed in Section 1.3.
Principle 1.1 ().
For any , if , then there exists a sequence in with .
The local equivalence between sequential and ‘epsilon-delta’ continuity cannot be proved in ZF, but can be established in (see [kohlenbach2]). Thus, it should not be a surprise that the latter system is often used as a base theory too.
Secondly, we make use the following standard definitions concerning sets.
Definition 1.2 (Sets).
-
•
A subset is given by its characteristic function , i.e. we write for , for any .
-
•
A set is enumerable if there is a sequence of reals that includes all elements of .
-
•
A set is countable if there is that is injective on , i.e.
-
•
A set is strongly countable if there is that is injective and surjective on ; the latter means that .
-
•
A set is finite in case there is such that for any finite sequence , there is with . We sometimes write ‘’.
Thirdly, we list the following second-order system needed below.
Principle 1.3 (finite--AC0, [gohzeg]).
The system plus for any arithmetical :
where ‘’ means that there is a non-empty sequence such that for any , .
We let height--AC0 be finite--AC0 where we additionally assume to be given such that for all , where is the length of the sequence in the formula ‘’. We have the following straightforward connections:
i.e. height--AC0 is also a system of hyperarithmetical analysis by Section 1.3. In the grand scheme of things, is a height function, a notion that goes back to Borel ([opborrelen3, opborrelen5]) and is studied in RM in [samBIG, samBIG3].
1.3. On hyperarithmetical analysis
Going back to Kreisel ([kreide]), the notion of hyperarithmetical set (see e.g. [simpson2]*VIII.3) gives rise to the second-order definition of theory/theorem of hyperarithmetical analyis (THA for brevity, see e.g. [skore3]). In this section, we recall known results regarding THAs, including the exact (rather technical) definition, for completeness.
First of all, well-known THAs are - and weak-- (see [simpson2]*VII.6.1 and VIII.4.12), where the latter is the former with the antecedent restricted to unique existence. Any system between two THAs is also a THA, which is a convenient way of establishing that a given system is a THA.
Secondly, at the higher-order level, from Section 1.2 is a conservative extension of - by [hunterphd]*Cor. 2.7. This is established by extending any model of -AC0 to a model of , where the second-order part of is isomorphic to . In this paper, we study (higher-order) systems that imply weak-- and are implied by . In light of the aforementioned conservation result, it is reasonable to refer to such intermediate third-order systems as existing in the range of hyperarithmetical analysis.
Thirdly, finding a natural THA, i.e. hailing from the mathematical mainstream, is surprisingly hard. Montalbán’s INDEC from [monta2], a special case of Jullien’s [juleke]*IV.3.3, is generally considered to be the first such statement. The latter theorem by Jullien can be found in [aardbei]*6.3.4.(3) and [roosje]*Lemma 10.3. The monographs [roosje, aardbei, juleke] are all ‘rather logical’ in nature and INDEC is the restriction of a higher-order statement to countable linear orders in the sense of RM ([simpson2]*V.1.1), i.e. such orders are given by sequences. In [dagsamXI]*Remark 2.8 and [samcount]*Remark 7 and §3.4, a number of third-order statements are identified, including the Bolzano-Weierstrass theorem and König’s infinity lemma, that are in the range of hyperarithmetical analysis. Shore and others have studied a considerable number of THAs from graph theory [skore1, skore2, gohzeg]. A related concept is that of almost theorem/theory of hyperarithmetical analysis (ATHA for brevity, [skore3]), which is weaker than but becomes a THA when combined with the latter.
Finally, we consider the official definition of THA from [monta2] based on -models.
Definition 1.4.
A system of axioms of second-order arithmetic is a theory/theorem of hyperarithmetical analysis in case
-
•
holds in for every , where is the -model consisting of all sets hyperarithmetic in ,
-
•
all -models of are hyperarithmetically closed.
Here, an -model is hyperarithmetically closed if it is closed under disjoint union and for every set , if is hyperarithmetically reducible to and is in the model, then is in the model too. In turn, this notion of reducibility means that can be expressed by a -formula with as a parameter; we refer to [monta2]*Theorem 1.14 for equivalent formulations.
2. Metric spaces
We introduce the well-known definition of metric space to be used in this paper (Section 2.1), where we always assume to be a subset of , up to coding of finite sequences. We show that basic properties of (Lipschitz) continuous functions on such metric spaces exist in the range of hyperarithmetical analysis (Section 2.2), even if we restrict to arithmetically defined objects (Theorem 2.8). We have previously studied metric spaces in [sammetric]; to our own surprise, some of these results have nice generalisations relevant to the study of hyperarithmetical analysis.
2.1. Basic definitions
We shall study metric spaces as in Definition 2.1. We assume that comes with its own equivalence relation ‘’ and that the metric satisfies the axiom of extensionality relative to ‘’ as follows:
Similarly to functions on the reals, ‘’ denotes a function from to the reals that satisfies the following instance of the axiom of function extensionality:
(EM) |
We recall that the study of metric space in second-order RM is at its core based on equivalence relations, as discussed explicitly in e.g. [simpson2]*I.4 or [damurm]*§10.1.
Definition 2.1.
A functional is a metric on if it satisfies the following properties for :
-
(a)
,
-
(b)
-
(c)
.
We shall only study metric spaces with or . To be absolutely clear, quantifying over amounts to quantifying over or , perhaps modulo coding of finite sequences, i.e. the previous definition can be made in third-order arithmetic for the intents and purposes of this paper. Since we shall study compact metric spaces, this restriction is minimal in light of [buko]*Theorem 3.13.
Sub-sets of are defined via characteristic functions, like for the reals in Definition 1.2, keeping in mind (EM). In particular, we use standard notation like to denote the open ball .
Secondly, the following definitions are now standard, where we note that a different nomenclature is sometimes used in second-order RM. A sequence in is Cauchy if
Definition 2.2 (Compactness and around).
For a metric space , we say that
-
•
is weakly countably-compact if for any in and sequence of rationals such that we have , there is such that ,
-
•
is countably-compact if for any sequence of open sets in such that , there is such that ,
-
•
is compact in case for any , there are such that covers ,
-
•
is sequentially compact if any sequence has a convergent sub-sequence,
-
•
is limit point compact if any infinite set in has a limit point,
-
•
is complete in case every Cauchy sequence converges,
-
•
is totally bounded if for all , there are such that covers .
-
•
a function is topologically continuous if for any open , the set is also open.
-
•
a function is closed if for any closed , we have that is also closed. ([munkies, ooskelly, leelee, searinghot]).
Regarding the final item, the set does not necessarily exist in , but ‘ is closed’ makes sense111In particular, ‘’ means ‘’ and ‘ is closed’ means ‘’, as expected. as shorthand for the associated well-known definition. We could study other notions, e.g. the Lindelöf property or compactness based on nets, but have opted to stick to basic constructs already studied in second-order RM.
Finally, fragments of the induction axiom are sometimes used, even in an essential way, in second-order RM (see e.g. [neeman, skore3]). The equivalence between induction and bounded comprehension is also well-known in second-order RM ([simpson2]*X.4.4). We shall need a little bit of the induction axiom as follows.
Principle 2.3 ().
Let satisfy . Then .
We let and be with ‘’ restricted to respectively ‘’ and ‘’. We have previously used for in the RM of the Jordan decomposition theorem ([dagsamXI]). By the proof of [dagsamXI]*Theorem 2.16, cannot prove the uncountability of the reals formulated as: the unit interval is not strongly countable.
2.2. Metric spaces and hyperarithmetical analysis
2.2.1. Introduction
In this section, we identify a number of the basic properties of metric spaces in the range of hyperarithmetical analysis, as listed on the next page. The Axiom of Choice for finite sets as in Principle 2.4 naturally comes to the fore. Clearly, the principle Finite Choice implies finite--AC0 over .
Principle 2.4 (Finite Choice).
Let be a sequence of non-empty finite sets in . Then there is such that for all .
In more detail, we will establish that the following theorems are intermediate between and .
-
•
Basic properties of continuous functions on sequentially compact metric spaces (Section 2.2.2).
-
•
Basic properties of sequentially continuous functions on (countably) compact metric spaces (Section 2.2.3).
-
•
Restrictions of the previous results to arithmetically defined or Lipschitz continuous functions (Section 2.2.4).
-
•
Basic properties of connected metric spaces, including the generalisation of the intermediate value theorem (Section 2.2.5).
We sometimes obtain elegant equivalences, like for the intermediate value theorem (Theorem 2.13). We believe there is no ‘universal’ approach to the previous results: each section is based on a very particular kind of metric space.
2.2.2. Sequentially compact spaces
In this section, we establish that basic properties of sequentially compact spaces inhabit the range of hyperarithmetical analysis. The following theorem is our first result, to be refined below.
Theorem 2.5 ().
The principle Finite Choice follows from any of the items (a)-(j) where is any metric space with ; the principle implies items (a)-(i).
-
(a)
For sequentially compact , any continuous is bounded.
-
(b)
The previous item with ‘is bounded’ replaced by ‘is uniformly continuous’.
-
(c)
For sequentially compact and continuous with given, there is with .
-
(d)
(Dini). Let be sequentially compact and let be a monotone sequence of continuous functions converging to continuous . Then the convergence is uniform.
-
(e)
For a sequentially compact metric space , equicontinuity implies uniform equicontinuity ([magnus]*Prop. 4.25).
-
(f)
For a sequentially compact metric space with infinite, there is a discontinuous function .
-
(g)
(Closed map lemma, [munkies, leelee, kura, mannetti]) For a sequentially compact metric space any continuous function is closed.
-
(h)
For sequentially compact and disjoint closed , .
-
(i)
(weak Cantor intersection theorem) For a sequentially compact metric space and a sequence of closed sets with , such that , there is a unique .
-
(j)
(Ascoli-Arzelà) For sequentially compact , a uniformly bounded equi-continuous sequence of functions on has a uniformly convergent sub-sequence.
The theorem still goes through if we require a modulus of continuity in item (a) or if we replace ‘continuity’ by ‘topological continuity’ in items (a)-(f).
Proof.
We first derive Finite Choice from item (a) via a proof-by-contradiction. To this end, fix a sequence of non-empty finite sets of reals . Suppose there is no sequence of reals such that for all . We now define
(2.1) |
where is a finite sequence of reals of length , readily coded using . We define the equivalence relation ‘’ as follows: the relation holds if , where . The metric is defined as for any . We then have as required. We also have for any , while for any we observe:
by the triangle equality of the absolute value on the reals. Hence, is a metric space as in Definition 2.1.
To show that is sequentially compact, let be a sequence in and consider the following case distinction. In case for some fixed , then contains at most different elements. The pigeon hole principle now implies that at least one occurs infinitely often in , i.e. is a convergent sub-sequence. In case , the sequence yields a sequence such that for all , which is impossible by assumption. Hence, is a sequentially compact metric space.
Next, define as , which is clearly not bounded on , which one shows using . To show that is continuous at , consider the formula ; the latter is false for and any . Hence, the following formula is vacuously true:
(2.2) |
i.e. is continuous at , with a modulus of continuity given by . To see that is also topologically continuous, fix an open set and fix . Then for , one verifies that , i.e. is open. Thus, is a continuous but unbounded function on a sequentially compact metric space , contradicting item (a). Item (b) also implies Finite Choice as is not uniformly continuous. For item (c), defined as is continuous in the same way as for . However, using , the infimum of on is , but there is no with , by definition. Hence, item (c) also implies Finite Choice.
Now assume item (d) and suppose Finite Choice is again false; letting and be as in the previous paragraph, we define as:
(2.3) |
Clearly, and for ; is continuous in the same way as for . Item (d) implies that the convergence is uniform, i.e.
(2.4) |
which yields a contradiction by letting be as in (2.4) for and choosing of length using . One derives Finite Choice from item (e) in the same way.
Next, regarding item (g), suppose Finite Choice is false and consider again . Define the continuous function by where is an enumeration of the rationals without repetitions. Using , we have and the latter is not closed while is, contradicting item (g), and Finite Choice must hold. To obtain the latter from item (f), note that is infinite (using ) while all functions are continuous as (2.2) is vacuously true. Regarding item (j), assuming again that Finite Choice is false, the sequence as in (2.3) is equicontinuous:
which (vacuously) holds in the same way as for (2.2). However, as for item (d), uniform convergence (of a sub-sequence) is false, i.e. item (j) also implies Finite Choice. For item (i), suppose Finite Choice is false, define , and verify that this closed and non-empty set has diameter at most , using . Since , we obtain Finite Choice from item (i). For item (h), suppose Finite Choice is false, and define and . One readily verifies that , are closed, and .
To establish the items in the theorem in , the usual proof-by-contradiction goes through. A proof sketch of item (a) as follows: let be as in the latter and suppose is continuous and unbounded, i.e. . Since and real numbers are represented by elements of Baire space, we may apply to obtain in such that for all . Since is sequentially compact, has a convergent sub-sequence, say with limit . Clearly, is not continuous at , a contradiction. To obtain (f), apply to the statement that is infinite, yielding a sequence in . Now define as and for for all . Since is unbounded on , it is discontinuous by item (a). Most other items are established using in the same way.
We also sketch how implies item (g). To this end, let be as in the closed map lemma and suppose is not closed for closed . Hence, there is such that . By definition, the latter formula means . Apply to obtain a sequence in with . By sequential compactness, there is a convergent sub-sequence , say with limit . Since is closed, we have and since is continuous (and hence sequentially continuous) also . This contradicts and the closed map lemma therefore follows from . ∎
The final part of the proof also goes through if is only usco (see Def. (4.1)). As to other generalisations of Theorem 2.5, the latter still goes through for ‘continuity’ replaced by ‘absolute differentiability’ from [incell] formulated222The correct formulation based on [incell] is that ‘ is (absolutely) differentiable on the metric space ’ in case we have which is the ‘epsilon-delta’ definition formulated to avoid the existence of the derivative. appropriately.
Finally, we observe that from (2.1) is not (countably) compact, i.e. we need a slightly different approach for the latter, to be found in the next section.
2.2.3. Compact spaces
In this section, we establish that basic properties of (countably) compact spaces inhabit the range of hyperarithmetical analysis.
First of all, the following theorem is a version of Theorem 2.5 for (countably) compact spaces and sequential continuity. We seem to (only) need sequential compactness to guarantee that everything remains provable in .
Theorem 2.6 ().
The principle Finite Choice follows from any of the items (a)-(d) where is any metric space with ; the principle implies all these items.
-
(a)
For (weakly) countably-compact and sequentially compact , any sequentially continuous is bounded.
-
(b)
The previous item with ‘is bounded’ replaced by ‘is (uniformly) continuous’.
-
(c)
For a (weakly) countably-compact that is infinite, there is that is not sequentially continuous.
-
(d)
The first item with ‘(weakly) countably-compact’ replaced by ‘compact’ or ‘complete and totally bounded’.
Proof.
We first derive Finite Choice from item (a) via a proof-by-contradiction. To this end, fix a sequence of non-empty finite sets of reals . Suppose there is no sequence of reals such that for all and recall from (2.1). Now define where is a new symbol such that for and ‘’ is ‘’ otherwise. Define as on , as for , and . Then is a metric, which is shown in the same way as for .
To show that is countably-compact, let be an open cover of and suppose is sucht hat . By definition, there is such that , i.e. implies for . Now use to enumerate the finitely many such that . This finite sequence is covered by some , i.e. we have obtained a finite sub-covering of , namely . Moreover, is sequentially compact, which can be proved via the same case distinction as for in the proof of Theorem 2.5.
Next, define as for and , which is clearly not bounded on ; this follows again via . Then is continuous at in the same way as from the proof of Theorem 2.5, namely since (2.2) is vacuously true. To show that is sequentially continuous at , let be a sequence converging to . In case this sequence is eventually constant , clearly as required. In case is not eventually constant , the convergence to in the -metric implies that for any , there is with . Thus, yields a sequence such that for all , which contradicts our assumptions, i.e. this case cannot occur. As a result, is sequentially continuous. Since, it is also unbounded (thanks to ), we obtain a contradiction with item (a). Thus, (a) implies Finite Choice, and the same for item (b). To obtain Finite Choice from item (c), note that is infinite (using ) while all functions are sequentially continuous by the previous.
To show that satisfies the properties in item (d), note that for , the ball covers all but finitely many points of (in the same way as from the second paragraph of the proof). Hence, is compact, and totally boundedness follows in exactly the same way. For completeness, let be a Cauchy sequence in , i.e. we have
As above, is either eventually constant or provides a sequence such that for all . The latter case is impossible by assumption and the former case is trivial.
To establish the items in the theorem in , the usual proof-by-contradiction goes through as in the proof of Theorem 2.5. ∎
We believe that we cannot use epsilon-delta or topological continuity in the previous theorem. Nonetheless, we have the following corollary that makes use of the sequential333A function is called sequentially uniformly continuous if for any sequences , in such that , we have . definition of uniform continuity.
Corollary 2.7.
Proof.
The usual proof-by-contradiction using (and ) shows that sequential uniform continuity implies uniform continuity. For the remaining implications, consider from the the proof of Theorem 2.6. This function is sequentially continuous at since any sequence converging to must be eventually constant . Similarly, for sequences , in implies that the sequences are eventually equal. Hence, is also sequentially uniformly continuous. A similar proof goes through for and from Theorem 2.5. ∎
We have identified a number of basic properties of continuous functions on compact metric spaces that exist in the range of hyperarithmetical analysis. A number of restrictions and variations are possible, which is the topic of the next section.
2.2.4. Restrictions
We show that some the above principles still inhabit the range of hyperarithmetical analysis if we restrict to arithmetically defined objects or Lipschitz continuity.
First of all, the following theorem establishes that Theorem 2.5 holds if we restrict to arithmetically defined objects.
Theorem 2.8.
Proof.
In a nutshell, we can modify the above proofs to obtain (only) weak--AC0 while all relevant objects can be defined using . To this end, let be arithmetical and such that , but there is no sequence with . Use to define such that outputs the binary expansions of , taking if there is only one. Define the following set using :
and . Then is injective and surjective on . In particular contains exactly one element by definition. Using , the metric space as in (2.1) in the proof of Theorem 2.5 now has an arithmetical definition. The same holds for the function where . The rest of the proof of item (a) of Theorem 2.5 now goes through, using instead of where relevant, yielding in particular a contradiction. Hence, there must be a sequence with , i.e. weak--AC0 follows as required. ∎
Secondly, we show that we may replace ‘continuity’ by ‘Lipschitz continuity’ in some of the above principles.
Definition 2.9.
A function is -Hölder-continuous in case there exist such that for any :
A function is Lipschitz (continuous) if is is -Hölder-continuous.
Theorem 2.10 ().
The principle Finite Choice follows from any of the items (a)-(e) where is any metric space with ; the principle implies items (a)-(e).
-
(a)
For a metric space , any sequentially compact is bounded, i.e. there are with (see [bartle2]*p. 333).
-
(b)
For sequentially compact , any uniformly continuous is bounded.
-
(c)
The previous item with ‘uniformly’ replaced by ‘-Hölder’ or ‘Lipschitz’.
-
(d)
For sequentially compact that is infinite, there exists that is bounded but not Lipschitz continuous.
-
(e)
For sequentially compact and bounded and Lipschitz with given, there is with .
Proof.
First of all, to derive item (a) from , fix a metric space and let be sequentially compact. Suppose is not bounded, i.e. for some fixed , we have . Apply to obtain a sequence such that for all . Clearly, this sequence cannot have a convergent sub-sequence, a contradiction, and must be bounded. To derive item (d) from , apply to the statement that is infinite. The resulting sequence has a convergent sub-sequence, say with limit . Define (resp. ) if and is even (resp. odd), and otherwise. Clearly, is bounded but not (Lipschitz) continuous. By, Theorem 2.5, the other items follow from .
Secondly, to derive Finite Choice from item (a), suppose is a sequence of finite sets such that there is no sequence with for all . Recall the set from (2.1) and define as for . That is a metric is readily verified: the first and third item of Definition 2.1 hold by definition and the triangle equality of the absolute value; the second item in this definition holds since . Now, the set is sequentially compact, as every sequence in either has at most finitely many different members, or yields a sequence such that for all . We have excluded the latter by assumption, while the former trivially yields a convergent sub-sequence. Using , is however not bounded in , a contradiction, and item (a) implies Finite Choice.
Thirdly, to derive Finite Choice from the remaining items, let be as above and note that the latter is sequentially compact as in the previous paragraph. Now define as and observe that , i.e. is Lifschitz (and uniformly) continuous. However, shows that is not bounded, a contradiction, and items (b)-(c) imply Finite Choice. Similarly, item (d) implies Finite Choice as is such that every bounded function is automatically Lipschitz. Indeed, if for all , then the Lipschitz constant for can be taken to be .
Finally, to derive Finite Choice from item (e), suppose the former is false and consider again , which is trivially bounded due to the definition of . Now define as . This function is Lipschitz on as
However, has infimum equal to zero (using ) but is strictly positive on , contradicting item (e), which establishes the theorem. ∎
In conclusion, many implications between the notions in Definition 2.2 exist in the range of hyperarithmetical analysis, as well as the associated Lebesgue number lemma for countable coverings of open sets. These are left to the reader.
2.2.5. Connectedness
We show that basic properties of connected metric spaces exist in the range of hyperarithmetical analysis, including the intermediate value theorem. We also obtain some elegant equivalences in Theorem 2.13.
First of all, Cantor and Jordan were the first to study connectedness ([wilders]), namely as in the first item in Definition 2.11. The connectedness notions from the latter are equivalent for compact metric spaces in light of [mannetti]*§4.39 or [pugh]*p. 123.
Definition 2.11 (Connectedness).
-
•
A metric space is chain connected in case for any and , there is a sequence such that for all we have .
-
•
A metric space is connected in case is not the disjoint union of two non-empty open sets.
We shall study the following generalisation of the intermediate value theorem.
Principle 2.12 (Intermediate Value Theorem).
Let be a sequentially compact and chain connected metric space and let be continuous. If for some and , then there is with .
The approximate intermediate value theorem is the previous principle with the conclusion weakened to ‘then for any there is with .’ The latter theorem is well-known from constructive mathematics (see e.g. [bridge1]*p. 40).
Theorem 2.13 ().
The principle Finite Choice follows from any of the items (a)-(g) where is any metric space with ; the principle implies items (a)-(g).
-
(a)
The intermediate value theorem as in Principle 2.12.
-
(b)
Principle 2.12 for Lipschitz continuous functions.
-
(c)
The approximate intermediate value theorem.
-
(d)
Let be a sequentially compact and chain connected metric space and let be continuous. Then is constant on .
-
(e)
Let be a sequentially compact and chain connected metric space and let be locally constant. Then is constant on .
-
(f)
Let be sequentially compact and chain connected and let be locally constant and continuous. Then is constant on .
-
(g)
For a sequentially compact metric space , chain connectedness implies connectedness
-
(h)
Let be a sequentially compact and chain connected metric space and let be (Lipschitz) continuous. Then is bounded on .
-
(i)
Item (h) with ‘ is bounded’ replaced by ‘ is not dense in ’.
-
(j)
Item (h) with ‘ is bounded’ replaced by ‘ is closed’.
Proof.
First of all, we show that item (a) implies Finite Choice. To this end, suppose the latter is false and consider as in (2.1). Let be an enumeration of the rationals (without repetitions) and define as follows: for . Then is a sequentially compact metric space, which is proved in the same way as for the previous metrics , namely that any sequence in can have at most finitely many different elements. That is chain connected is proved using . Indeed, fix and consider . Let be a finite sequence such that for . Using , there are such that for , and chain connectedness of follows.
Now define by , which is (Lipschitz) continuous, essentially by the definition of , as we have:
However, the range of consists of rationals, i.e. it does not have the intermediate value property. This contradiction yields Finite Choice. The same proof goes through for items (h)-(j).
Secondly, assume and let and be as in Principle 2.12. Since is chain connected, we have
(2.5) |
Apply to obtain a sequence of finite sequences. Define a sequence in where is the first element in such that . By sequential completeness, there is a convergent sub-sequence with limit . Since is continuous, we have and hence .
Thirdly, to show that item (g) implies Finite Choice, again suppose the latter is false and consider . By the above, the latter is sequentially compact and chain connected. To show that it is not connected, define and , verify that they are open and disjoint, and observe that , i.e. item (g) is false. Note also that defined as if and otherwise, is continuous but not constant, i.e. item (d) also implies Finite Choice.
To derive item (g) from , let be as in the former, i.e. sequentially compact and chain connected. Suppose is not connected, i.e. where the latter are open, disjoint, and non-empty. Now fix and and consider (2.5). Apply to obtain a sequence of finite sequences. Define sequences and in where is the first element in such that and is the predecessor of in . By sequential completeness, and have convergent sub-sequences, with the same limit by construction. However, if this limit is in , then so is eventually, a contradiction. Similarly, if this limit is in , then so is eventually, a contradiction. In each case we obtain a contradiction, i.e. must be connected, and item (g) follows. The same proof goes through for item (d).
Next, item (g) implies item (e) as in case the latter fails for , say with , then and are open, disjoint, and non-empty sets such that , i.e. item (g) fails too. To show that item (e) implies Finite Choice, suppose the latter is false and let be as above. Define as in case . Clearly, is locally constant but not constant, i.e. item (e) is false. To derive item (g) from item (e) (and item (f)), suppose the former is false, i.e. is a sequentially compact and chain connected metric space that is not connected. Let be the associated decomposition and note that defined by if and otherwise, is locally constant (and continuous) but not constant, i.e. item (e) (and (f)) also fails. The equivalence for item (d) follows in the same way.
To show that item (g) implies item (a), suppose the latter is false for and , i.e. for all . By assumption, and are open, disjoint, and non-empty, i.e. item (g) also fails. To show that item (a) (and item (c)) implies item (g), suppose the latter fails for , i.e. the latter are open, non-empty, and disjoint. Then defined by if and otherwise, is continuous but does not have the (approximate) intermediate value property. ∎
Regarding item (i), we could not find a way of replacing ‘ is not dense in ’ by ‘ has finite measure’. We could study local connectedness and obtain similar results, but feel this section is long enough as is.
In conclusion, we have identified many basic properties of metric spaces that exist in the range of hyperarithmetical analysis. We believe there to be many more such principles in e.g. topology.
3. Functions of Bounded variation and around
We introduce functions of bounded variation (Section 3.1) and show that their basic properties exist in the range of hyperarithmetical analysis (Section 3.2). Similar to Theorem 2.8, we could restrict to arithmetically defined functions.
3.1. Bounded variation and variations
The notion of bounded variation (often abbreviated ) was first explicitly444Lakatos in [laktose]*p. 148 claims that Jordan did not invent or introduce the notion of bounded variation in [jordel], but rather discovered it in Dirichlet’s 1829 paper [didi3]. introduced by Jordan around 1881 ([jordel]) yielding a generalisation of Dirichlet’s convergence theorems for Fourier series. Indeed, Dirichlet’s convergence results are restricted to functions that are continuous except at a finite number of points, while -functions can have infinitely many points of discontinuity, as already studied by Jordan, namely in [jordel]*p. 230. In this context, the total variation of is defined as:
(3.1) |
The following definition provides two ways of defining ‘-function’. We have mostly studied the first one ([dagsamXI, samBIG, samBIG3]) but will use the second one in this paper.
Definition 3.1 (Variations on variation).
-
(a)
The function has bounded variation on if there is such that for any partition .
-
(b)
The function has total variation on if .
We recall the ‘virtual’ or ‘comparative’ meaning of suprema in RM from e.g. [simpson2]*X.1. In particular, a formula ‘’ is merely shorthand for (essentially) the well-known definition of the supremum.
Secondly, the fundamental theorem about -functions is formulated as follows.
Theorem 3.2 (Jordan decomposition theorem, [jordel]*p. 229).
A -function is the difference of two non-decreasing functions .
Theorem 3.2 has been studied via second-order representations in [groeneberg, kreupel, nieyo, verzengend]. The same holds for constructive analysis by [briva, varijo, brima, baathetniet], involving different (but related) constructive enrichments. We have obtained many equivalences for the Jordan decomposition theorem, formulated using item (a) from Definition 3.1 in [dagsamXI, samBIG3], involving the following principle.
Principle 3.3 ().
A countable set can be enumerated.
This principle is ‘explosive’ in that proves while proves (see [dagsamX]*§4).
Thirdly, is regulated if for every in the domain, the ‘left’ and ‘right’ limit and exist. Feferman’s readily provides the limit of if it exists, i.e. the notation for regulated makes sense in . On a historical note, Scheeffer and Darboux study discontinuous regulated functions in [scheeffer, darb] without using the term ‘regulated’, while Bourbaki develops Riemann integration based on regulated functions in [boerbakies]. Finally, -functions are regulated while Weierstrass’ ‘monster’ function is a natural example of a regulated function not in .
3.2. Bounded variation and hyperarithmetical analysis
We identify a number of statements about -functions that exist within the range of hyperarithmetical analysis, assuming . We even obtain some elegant equivalences and discus the (plentiful) variations of these results in Section 4.3.
First of all, the following principle appears to be important, which is just from the previous section restricted to strongly countable sets.
Principle 3.4 ().
A strongly countable set can be enumerated.
Some RM-results for may be found in [dagsamXI]*§2.2.1; many variations are possible and these systems all exist in the range of hyperarithmetical analysis. The cited results are not that satisfying as they mostly deal with properties of strongly countable sets, in contrast to the below.
Secondly, we have the following theorem, establishing that items (ii)-(v) exist in the range of hyperarithmetical analysis.
Theorem 3.5 ().
The higher items imply the lower ones.
-
(i)
The principle .
-
(ii)
(Jordan) For with , there are non-decreasing such that .
-
(iii)
For with , there is a sequence that includes all points of discontinuity of .
-
(iv)
For with , the supremum555To be absolutely clear, we assume, for the existence of a functional such that ). exists for .
-
(v)
.
-
(vi)
weak--AC0.
Items (ii)-(iii) are equivalent; we only use to derive from item (5).
Proof.
Assume and let be such that . By [dagsamXIV]*Theorem 2.16, suffices to enumerate all jump discontinuities of a regulated function, while is regulated by [dagsamXI]*Theorem 3.33. Then implies that
The formula in square brackets is arithmetical, i.e. since is available we may apply to obtain a sequence of finite sequences witnessing the previous centred formula. This sequence includes all removable discontinuities of . Indeed, suppose is such that is not among the reals in . Let be such that and note that for by assumption. Extending with and points close enough to , we obtain a partition of that witnesses that , contradicting our assumptions. Since is regulated, it only has removable and jump discontinuities, i.e. item (iii) follows from as required.
By [dagsamXI]*Theorem 3.33, suffices to enumerate the points of discontinuity of any monotone , i.e. item (ii) implies item (iii). To obtain item (ii) from item (iii), note that the supremum over in (3.1) can be replaced by a supremum over and any sequence that includes all points of discontinuity of . Hence, we may use to define the weakly increasing function . One readily verifies that is also weakly increasing, i.e. as in item (ii) follows. To obtain item (5) from item (iii), note that -similar to the previous- the supremum over in can be replaced by a supremum over and any sequence that includes all points of discontinuity of .
To derive from item (5), let and such that the latter is injective and surjective on the former. Now define as follows: if , and otherwise. Using , is in and . Now use to decide whether ; if the latter holds, ‘’ is the first bit of the binary expansion of such that . Using the supremum functional and , the usual interval-halving technique then allows us to enumerate , as required for . For the final part, let be arithmetical and such that . Use to define such that outputs the binary expansions of , taking if there is only one. Then is a singleton and is injective and surjective on . The enumeration of provided by yields the consequent of weak--AC0. ∎
As to the role of the Axiom of Choice in Theorem 3.5, we note that the items (ii)-(v) can also be proved without . Indeed, as in (3.1) involves a supremum over , which can be defined in using the well-known interval-halving technique, i.e. the usual textbook proof (see e.g. [voordedorst]) goes through in .
Thirdly, we have the following corollary using slightly more induction.
Corollary 3.6.
Proof.
We note that suffices to guarantee that -functions are regulated by [dagsamXI]*Theorem 3.33. Now, the Fourier series of a -function always converges to and this fact is provable in if the Fourier coefficients are given, as discussed in (a lot of detail in) [samBIG]*§3.4.4. Hence, item (iii) of Theorem 3.5 immediately implies the centred statement in item (a), while for the reversal, the centred statement provides a sequence that includes all removable discontinuities, i.e. where but . By [dagsamXIV]*Theorem 2.16, suffices to enumerate all jump discontinuities of a regulated function. Since there are no other discontinuities for , the corollary follows. ∎
We could obtain similar results for e.g. Bernstein or Hermit-Fejer polynomials as analogous results hold for -functions (see [samBIG3]). Other variations are discussed in Remark 4.3 below.
Fifth, as noted in Section 3.1, enumerating the points of discontinuity of a regulated function implies ; the latter yields when combined with . By contrast, item (ii) in the following theorem is much weaker.
Theorem 3.7 ().
The higher items imply the lower ones.
-
(i)
The principle .
-
(ii)
For regulated with infinite , there is a sequence of distinct points of discontinuity of .
-
(iii)
The principle Finite Choice.
-
(iv)
The principle finite--AC0.
Proof.
The first downward implication is immediate by applying -modulo - to ‘ is not finite’. The final implication is straightforward. For the second downward implication, let be a sequence of non-empty finite sets and let be such that is the binary expansion of , choosing a tail of zeros if necessary. Define as:
Using , one readily shows that is regulated (with left and right limits equal to zero) and that is infinite if is. Any sequence in then yields a sequence as in the consequent of Finite Choice. ∎
An interesting variation is provided by the following corollary. We conjecture that Finite Choice cannot be obtained from the second item.
Corollary 3.8 ().
The higher items imply the lower ones.
-
(i)
The principle .
-
(ii)
For in with infinite , there is a sequence of distinct points of discontinuity of .
-
(iii)
() Let be a sequence of non-empty finite sets in and let be such that . Then there is a sequence such that for all .
-
(iv)
The principle height--AC0.
Proof.
The final implication is straightforward while the first one follows as in the proof of the theorem. For the second downward implication, let be a sequence of non-empty finite sets with . Define as in the proof of the theorem but replacing ‘’ in the first case by . By construction, is in with and the set is infinite if is. Any sequence in then yields the sequence as in the consequent of . ∎
Finally, we discuss numerous possible variations of the above results in Section 4.3, including Riemann integration and rectifiability.
4. Other topics in hyperarithmetical analysis
4.1. Semi-continuity and closed sets
We show that basic properties of semi-continuous functions, like the extreme value theorem, exist in the range of hyperarithmetical analysis. Since upper semi-continuous functions are closely related to closed sets, the latter also feature prominently.
First of all, we need Baire’s notion of semi-continuity first introduced in [beren].
Definition 4.1.
For , we have the following definitions:
-
•
is upper semi-continuous at if for any , there is such that .
-
•
is lower semi-continuous at if for any , there is such that .
We use the common abbreviations ‘usco’ and ‘lsco’ for the previous notions. We say that ‘ is usco’ if is usco at every . Following [martino], the extreme value theorem does not really generalise beyond semi-continuous functions.
Secondly, we have the following theorem, a weaker version of which is in [dagsamXVI]. We repeat that since the characteristic function of a closed set is usco, the connection between items (ii) and ClC is not that surprising.
Theorem 4.2 ().
The higher items imply the lower ones.
-
(i)
The principle .
-
(ii)
(Extreme value theorem) For usco with for all , there is such that .
-
(iii)
(ClC, [dagsamXVI]) Let be a sequence of non-empty closed sets in . Then there is such that for all .
-
(iv)
For usco and regulated with for all , there is such that .
-
(v)
(Finite Choice) Let be a sequence of non-empty finite sets in . Then there is such that for all .
-
(vi)
The principle finite--AC0.
Proof.
For the first downward implication, if the supremum is given, we have , and applying yields a sequence . Since , the latter has a convergent sub-sequence (for fixed ), with limit say by sequential completeness. One readily verifies that for any as is usco. For the second implication, fix a sequence of closed sets and define as follows using Feferman’s :
(4.1) |
Since is essentially the characteristic function of closed sets, is usco on by definition, for each . The sequence provided by item (ii) then clearly satisfies . To show that ClC implies item (ii), let and be as in the latter and define which is non-empty by definition and closed as is usco. The sequence provided by ClC yields such that . The function from (4.1) is also regulated in case each is finite, i.e. the fourth implication also follows. ∎
We note that item (ii) is equivalent to e.g. the sequential version of the Cantor intersection theorem ([dagsamXVI]).
Thirdly, ClC from Theorem 4.2 is provable in if assume that the closed sets are given by a sequence of RM-codes (see [simpson2]*IV.1.8). We next study ClC for an alternative representation of closed sets from [browner, brownphd, browner2] as follows.
Definition 4.3.
A (code for a) separably closed set is a sequence of reals. We write ‘’ in case . A (code for a) separably open set is a code for the (separably closed) complement.
Next, item (i) in Theorem 4.4 is a weakening of [simpson2]*V.4.10, which in turn is a second-order version of the countable union theorem. In each case, the antecedent only expresses that for every , there exists an enumeration of ; abusing notation666In particular, the formula ‘’ in Theorem 4.4 is short-hand for which is slightly more unwieldy. slightly, we still write ‘’ as in Definition 4.3, leaving the enumeration of implicit. We sometimes identify subsets and elements .
Theorem 4.4 ().
The following items are intermediate between -AC0 and weak--AC0.
-
(i)
Let be a sequence of analytic codes such that each is enumerable and non-empty. There is a sequence with .
-
(ii)
Let be a sequence of analytic codes such that is enumerable and has positive measure. There exists with .
-
(iii)
Let be a sequence of analytic codes such that is enumerable and is not enumerable. There exists with .
-
(iv)
Let be a sequence of analytic codes such that for all , is RM-open. There exists with .
Proof.
To prove the items in -AC0, apply the latter to , noting that the formula in square brackets is if is an analytic code. To derive weak--AC0 from item (i), let be arithmetical and such that and define ‘’ as using [simpson2]*V.1.7′. Clearly, then implies as codes a singleton, i.e. item (i) implies weak--AC0. To obtain weak--AC0 from item (iv), let be as in the antecedent of the former and consider defined as
(4.2) |
which yields a sequence of -formulas, yielding in turn a sequence of analytic codes by [simpson2]*V.1.7′. In light of (4.2), is a basic open ball in . In case for all , define and note that for all . To obtain weak--AC0 from item (ii), let be as in the antecedent of the former and define as
which yields a sequence of -formulas, yielding in turn a sequence of analytic codes by [simpson2]*V.1.7′. For fixed , there is a unique such that , immediately yielding an enumeration of for any . Essentially by definition, has measure . In case for all , define and note that for all . Item (iii) also follows as enumerable sets have measure zero. ∎
We would like to formulate item (i) using Borel codes from [simpson2]*V.3, but the latter seem to need to express basic aspects. The items from the theorem also imply finite--AC0, which is left as an exercise.
Finally, we formulate a higher-order result for comparison; we continue the abuse of notation involving as in Theorem 4.2.
Theorem 4.5 ().
The higher items imply the lower ones.
-
(i)
The principle .
-
(ii)
Let be a sequence of sets in such that for all , is enumerable and non-empty. There is with .
-
(iii)
Let be a sequence of sets in such that for all , is enumerable and has positive measure. There is with .
-
(iv)
Proof.
The first downward implication follows by applying to ‘ is non-empty for all ’. For the third downward implication, let and be such that . Define the set
and note that this sequence has a straightforward enumeration while the associated separably closed set has measure . Let be the sequence provided by item (iii). Using sequential compactness, is a real in satisfying , for any as required. ∎
Variations of the previous theorem are possible, e.g. replacing ‘enumerable’ by ‘(strongly) countable’. Nonetheless, we are not able to derive e.g. from ClC restricted to closed sets of positive measure, i.e. the previous two theorems may well be due to the coding of closed sets as in Definition 4.3.
4.2. Unordered sums
The notion of unordered sum is a device for bestowing meaning upon sums involving uncountable index sets. We introduce the relevant definitions and then prove that basic properties of unordered sums exist in the range of hyperarithmetical analysis.
First of all, unordered sums are essentially ‘uncountable sums’ for any index set and . A central result is that if somehow exists, it must be a ‘normal’ series of the form , i.e. for all but countably many ; Tao mentions this theorem in [taomes]*p. xii.
By way of motivation, there is considerable historical and conceptual interest in this topic: Kelley notes in [ooskelly]*p. 64 that E.H. Moore’s study of unordered sums in [moorelimit2] led to the concept of net with his student H.L. Smith ([moorsmidje]). Unordered sums can be found in (self-proclaimed) basic or applied textbooks ([hunterapp, sohrab]) and can be used to develop measure theory ([ooskelly]*p. 79). Moreover, Tukey shows in [tukey1] that topology can be developed using phalanxes, which are nets with the same index sets as unordered sums.
Secondly, as to notations, unordered sums are just a special kind of net and is therefore written in this context to suggest the connection to nets. The associated notation is purely symbolic. We only need the following notions in the below. Let be the set of all finite sequences of reals without repetitions.
Definition 4.6.
Let be any mapping, also denoted .
-
•
We say that is convergent to if for all , there is such that for with , we have .
-
•
A modulus of convergence is any sequence such that for all in the previous item.
For simplicity, we focus on positive unordered sums, i.e. such that for .
Thirdly, we establish that basic properties of unordered sums exist in the range of hyperarithmetical analysis.
Theorem 4.7 ().
The higher items imply the lower ones.
-
(i)
.
-
(ii)
For a positive and convergent unordered sum , there is a sequence of reals such that for all not in this sequence.
-
(iii)
For a positive and convergent unordered sum , there is a modulus of convergence.
-
(iv)
.
Proof.
Assume and note that the convergence of an unordered sum to some implies
(4.3) |
Apply to (4.3) to obtain a sequence of finite sequences of reals. This sequence must contain all such that . Indeed, suppose satisfies for fixed and is not included in . By definition, satisfies . However, for , we have , a contradiction. Hence, implies item (ii). The second and third items are readily seen to be equivalent.
For the final downward application, let and be such that the latter is injective and surjective on the former. Define if , and otherwise. One readily proves that is convergent to , for which is needed. The sequence from the second item now yields the enumeration of the set required by . ∎
We note that height--AC0 can be obtained from item (ii) in Theorem 4.7; we conjecture that finite--AC0 cannot be obtained. Since unordered sums are just nets, one could study statements like
a convergent net has a convergent sub-sequence,
which for index sets defined over Baire space is equivalent to ([samhabil]).
4.3. Variations and generalisations
We discuss variations and generalisations of the above results.
First of all, many variations of the results in Section 3.2 exist for rectifiable functions. Now, Jordan proves in [jordel3]*§105 that -functions are exactly those for which the notion of ‘length of the graph of the function’ makes sense. In particular, if and only if the ‘length of the graph of ’, defined as follows:
(4.4) |
exists and is finite by [voordedorst]*Thm. 3.28.(c). In case the supremum in (4.4) exists (and is finite), is also called rectifiable. Rectifiable curves predate -functions: in [scheeffer]*§1-2, it is claimed that (4.4) is essentially equivalent to Duhamel’s 1866 approach from [duhamel]*Ch. VI. Around 1833, Dirksen, the PhD supervisor of Jacobi and Heine, already provides a definition of arc length that is (very) similar to (4.4) (see [dirksen]*§2, p. 128), but with some conceptual problems as discussed in [coolitman]*§3.
Secondly, regulated functions are not necessarily but have bounded Waterman variation (see [voordedorst]), which is a generalisation of where the sum in (3.1) is weighted by a Waterman sequence, which is a sequence of positive reals that converges to zero and with a divergent series. Some of the above results generalise to regulated function for which the Waterman variation is known, say .
Thirdly, one can replace the consequent of item (iii) in Theorem 3.5 by a number of similar conditions, like the existence of a Baire 1 representation (which can be defined in for monotone functions), the fundamental theorem of calculus at all reals but a given sequence, or the condition that if the Riemann integral of in is zero, for all but a given sequence. Many similar conditions may be found in [samBIG, samBIG3, samBIG4].
Fourth, Theorem 3.7 is readily generalised to (almost) arbitrary functions on the reals. To make sure the resulting theorem is provable in , it seems we need oscillation functions777For any , the associated oscillation functions are defined as follows: and . Riemann, Ascoli, and Hankel already considered the notion of oscillation in the study of Riemann integration ([hankelwoot, rieal, ascoli1]), i.e. there is ample historical precedent. In the same way as for Theorem 3.7, one proves that the higher items imply the lower ones over .
-
•
The principle .
-
•
Any infinite set has a limit point.
-
•
For any with oscillation function , the set is either finite or has a limit point.
-
•
For a non-piecewise continuous with oscillation function , the set has a limit point.
-
•
The arithmetical Bolzano-Weierstrass theorem ([coniving]).
We note that is necessary to make ‘’ into an arithmetical formula while ‘ is a limit point of ’ is a meaningful (non-arithmetical) formula even if does not exist as a set.
Acknowledgement 4.8.
Our research was supported by the Klaus Tschira Boost Fund via the grant Projekt KT43. The initial ideas for this paper, esp. Section 2, were developed in my 2022 Habilitation thesis at TU Darmstadt ([samhabil]) under the guidance of Ulrich Kohlenbach. The main ideas of this paper came to the fore during the Trends in Proof Theory workshop in February 2024 at TU Vienna. We express our gratitude towards all above persons and institutions.
References
- \bibselectallkeida