I am currently following the book Computational Type Theory and Interactive Theorem Proving with Coq by Gert Smolka and on page 75, there is exercise 9.3.14.b asking me to prove that the 'weak law of excluded middle' defined as:
Definition WXM : Prop := forall (X:Prop), ~X \/ ~~X.
implies the Morgan law for conjunction:
Definition MGC : Prop := forall (X Y:Prop), ~(X/\Y) -> ~X \/ ~Y. (* <- always true *)
i.e.
Lemma L1 : WXM -> MGC
Proof.
Admitted.
I have been trying for solve this for a while now, but with no success. Assuming WXM
and ~(X/\Y) for some Prop X Y, faced with the goal ~X \/ ~Y, I did a case
analysis (applying WXM on X and Y). Out of the 4 cases, 3 are immediately dispatched but I am left with a 4th cases with additional assumptions ~~X and ~~Y.
Intuitively ~~X and ~~Y are saying that X and Y are 'weakly true' and you would hope to conclude that X /\ Y is also weakly true (i.e. show ~~(X/\Y) leading to a contradiction thanks to the assumption ~(X/\Y)). However, I am not able to conclude.
I don't want to give up but would also like to move on with the book. Does anyone have the answer to this?.
One does not need to consider the four cases here, because one has negated disjuncts in the goal giving some terms to work with when one chooses which disjunct to prove. Getting e.g. ~ Q and ~ ~ Q cases is enough to prove the lemma.
I'm not sure how to explain this intuition further without just showing the solution.
Goal
(forall P : Prop, ~ P \/ ~ ~ P) ->
(forall P Q : Prop, ~ (P /\ Q) -> ~ P \/ ~ Q).
Proof.
intros wlem P Q npq.
destruct (wlem Q) as [nq | nnq].
- right; trivial.
- left; intros p.
apply nnq; intros q.
apply npq; split; trivial.
Qed.
Related
~ (exists x:D, ~ R x)->(forall y:D, R y)
I have worked on it for quite a long time, but it seems that I cannot use the left part of the implication well.
This is the first part of my code:
Parameter D: Set.
Parameter x: D.
Parameter y: D.
Parameter R: D->Prop.
Lemma b: ~(exists x:D, ~ (R x))->(forall y:D, (R y)).
Can anyone help me figure out how to write the rest of of the code?
Your question is a bit vague, as you don't specify what D and R are, and where you are stuck in your proof. Try providing a minimal working example, with an explicit fail tactic for where you're stuck in the proof.
In classical logic (the one you're use to in maths), as you have the excluded middle rule, you can always do a case analysis on whether something is true or false. In vanilla Coq, built for intuitionistic logic, it's not the case. Your result is actually not provable if the predicate R is not decidable (if it's not either true or false on every input : forall (x:D), R x \/ ~R x), if the type D is not empty.
Try adding the decidability of R as an hypothesis and reprove it. It should follow more or less this structure (the key being the case analysis on whether (R y) is true or false) :
Parameter D: Set.
Parameter R: D -> Prop.
Lemma yourGoal :
(forall x, R x \/ ~ R x) -> (* Decidability of R *)
~ ( exists x, ~ (R x) )->
forall y, (R y).
Proof.
intros Hdec Hex y. (* naming the hypothesis for convenience *)
specialize (Hdec y).
destruct Hdec as [H_Ry_is_true | H_Ry_is_false]. (* case analysis, creates two goals *)
+ (* (R y) is true, which is our goal. *)
assumption.
+ (* (R y) is false, which contradicts Hex *)
exfalso. (* transform your goal into False *)
apply Hex.
(* should be easy from here, using the [exists] tactic *)
Qed.
Ps: this exact result (and its link with excluded middle) is mentioned in Software foundations, which is a great resource to learn Coq and logic : https://softwarefoundations.cis.upenn.edu/lf-current/Logic.html#not_exists_dist
I am a beginner at coq.
I do not know the meaning of intros [=] and intros [= <- H] . and I could not find an easy explanation. Would someone explain these two to me please?
Regards
The documentation for this is here. I will add a little explanation note.
The first historical use of intro patterns is to decompose data that is packed in inductive objects on the fly. Here is a first easy example (tested with coq 8.13.2).
Lemma forall A B, A /\ B -> B /\ A.
Proof.
If you run the tactic intros A B H then the hypothesis H will be a proof of A /\ B. Morally, this contains knowledge that A holds, but it cannot be used as such, because it is a proof of a stronger fact. It is often the case that users want directly to decompose this hypothesis, this would normally be done by typing destruct H as [Ha Hb]. But if you know right away that you will not keep hypothesis H, why not find a shorter expression. This is what the intro pattern is used for.
So you type the following command and have the resulting goal:
Intros A B [Ha Hb].
(* resulting goal
A, B : Prop
Ha : A
Hb : B
============================
B /\ A
*)
Abort.
I will not finish this proof. But you get the idea of what intro patterns are for: decompose information on the fly when inductive types (like conjunction here) pack several pieces of information together.
Now, equality information also can pack several pieces of information together. Assume now that we are working with lists of natural numbers and we have the following equality.
Require Import List.
Lemma intro_pattern_example2 n m p q l1 l2 :
(n :: S m :: l1) = (p :: S q :: l2) -> q :: p :: l2 = m :: n :: l1.
The equality in the left-hand side of the implication is an equality between two lists, but it actually packs several more elementary pieces of information: n = p, m = q, and l1 = l2. If you just type intros H, you obtain the equality between two lists of length 3, but if you type intros [=], you ask the proof system to explore the structure of each equality member and check when constructors appear so that the smaller pieces of information can be placed in separate hypothesis instead of the big one. This is a short hand for the use of the injection tactic. Here is the example.
intros [= Hn Hm Hl1].
(*resulting goal:
n, m, p, q : nat
l1, l2 : list nat
Hn : n = p
Hm : m = q
Hl1 : l1 = l2
============================
q :: p :: l2 = m :: n :: l1
*)
So you see, this intro pattern unpacks information that would otherwise be stuck in a more complex hypothesis.
Now, when an hypothesis is an equality, there is another action you might want to perform right away. You might want to rewrite with it. In intro patterns, this is done by replacing the name you would give to that equality with an arrow. Let's test this on the previous goal.
Undo.
intros [= -> -> ->].
(* resulting goal
p, q : nat
l2 : list nat
============================
q :: p :: l2 = q :: p :: l2
*)
Now this goal can be solved quickly with reflexivity, trivial, or auto. Please note that the hypotheses were used to rewrite, but they were not kept in the goal context, so this possibility to rewrite directly from the intro pattern has to be used with caution, because you are actually losing some information.
The [= ] intro pattern is used especially for equalities and when both members are datatype constructors. It exploits the natural injectivity property of these constructors. there is another property that is respected by datatype constructors. It is the fact that two pieces of data with different head constructors can never be equal. This is exploited in Coq by the discriminate tactic. The [=] intro pattern is shorthand for both the injection and discriminate tactics.
I wanted to see a few hands on examples of Coq proofs of the form:
\exists A(x1,...,xn)
essentially where the Goal had an existential quantifier. I was having issues manipulating the goal in meaningful ways to make progress in my proof and wanted to see a few examples of common tactics to manipulate.
What are some good existential quantifiers examples in Coq to prove?
My specific example I had:
Theorem Big_Small_ForwardImpl :
forall (P : Program) (S' : State),
(BigStepR (B_PgmConf P) (B_StateConf S')) -> (ConfigEquivR (S_PgmConf P) (S_BlkConf EmptyBlk S')).
Proof.
intros.
induction P.
unfold ConfigEquivR.
refine (ex_intro _ _ _) .
my context and goals was:
1 subgoal
l : list string
s : Statement
S' : State
H : BigStepR (B_PgmConf (Pgm l s)) (B_StateConf S')
______________________________________(1/1)
exists N : nat, NSmallSteps N (S_PgmConf (Pgm l s)) (S_BlkConf EmptyBlk S')
but then changed to:
1 subgoal
l : list string
s : Statement
S' : State
H : BigStepR (B_PgmConf (Pgm l s)) (B_StateConf S')
______________________________________(1/1)
NSmallSteps ?Goal (S_PgmConf (Pgm l s)) (S_BlkConf EmptyBlk S')
after using the refine (ex_intro _ _ _) tactic. Since I am not sure what is going on I was hoping some simpler examples could show me how to manipulate existential quantifiers in my Coq goal.
helpful comment:
The ?Goal was introduced by Coq as a placeholder for some N that will have to be deduced later in the proof.
The following example is based on the code provided in this answer.
Suppose we have a type T and a binary relation R on elements of type T. For the purpose of this example, we can define those as follows.
Variable T : Type.
Variable R : T -> T -> Prop.
Let us prove the following simple theorem.
Theorem test : forall x y, R x y -> exists t, R x t.
Here is a possible solution.
Proof.
intros. exists y. apply H.
Qed.
Instead of explicitly specifying that y is the element we are looking for, we can rely on Coq's powerful automatic proof mechanisms in order to automatically deduce which variable satisfies R x t:
Proof.
intros.
eexists. (* Introduce a temporary placeholder of the form ?t *)
apply H. (* Coq can deduce from the hypothesis H that ?t must be y *)
Qed.
There exist numerous tactics that make ise of the same automated deduction mechanisms, such as eexists, eapply, eauto, etc.
Note that their names often correspond to usual tactics prefixed with an e.
I am learning propositional logic and the rules of inference. The Disjunctive Syllogism rule states that if we have in our premises (P or Q), and also (not P); then we can reach Q.
I can not for the life of me figure out how to do this in Coq. Let's say I have :
H : A \/ B
H0 : ~ A
______________________________________(1/1)
What tactic should I use to reach
H1 : B.
As an extra, I would be glad if someone could share with me the Coq tactic equivalents of basic inference rules, like modus tollens, or disjunctive introduction etc. Is there maybe a plugin I can use?
Coq does not have this tactic built-in, but fortunately you can define your own tactics. Notice that
destruct H as [H1 | H1]; [contradiction |].
puts H1 : B in the context, just as you asked. So you can create an alias for this combined tactic:
Ltac disj_syllogism AorB notA B :=
destruct AorB as [? | B]; [contradiction |].
Now we can easily imitate the disjunctive syllogism rule like so:
Section Foo.
Context (A B : Prop) (H : A \/ B) (H0 : ~ A).
Goal True.
disj_syllogism H H0 H1.
End Foo.
Let me show a couple less automated approaches:
Ltac disj_syllogism AorB notA B :=
let A := fresh "A" in
destruct AorB as [A | B]; [contradiction (notA A) |].
This approach does not ask Coq to find a contradiction, it provides it directly to the contradiction tactic (notA A term). Or we could have used an explicit term with the pose proof tactic:
Ltac disj_syllogism AorB notA B :=
pose proof (match AorB with
| or_introl a => False_ind _ (notA a)
| or_intror b => b
end) as B.
I hope this helps. I'm not sure if some extra explanation is needed -- feel free to ask for clarification and I'll update my answer.
I think you maybe have the wrong expectations on how Coq works? The general way of proving this is essentially a truth-table on the different possibilities:
Lemma it: forall a b, (a \/ b) /\ ~a -> b.
Proof.
intuition.
Show Proof.
Qed.
(fun (a b : Prop) (H : (a \/ b) /\ ~ a) =>
and_ind
(fun (H0 : a \/ b) (H1 : ~ a) =>
or_ind (fun H2 : a => let H3 : False := H1 H2 in False_ind b H3)
(fun H2 : b => H2) H0) H)
If you look at the resulting proof-term, you see the Coq is essentially taking apart the boolean the constructors. We can do this manually and get the same proof-term:
Lemma it: forall a b, (a \/ b) /\ ~a -> b.
Proof.
intros a b H.
induction H.
induction H.
contradict H. exact H0.
exact H.
Qed.
Whereas e.g. modus ponens corresponds to an apply in Coq, I don't think this is "built in" in any direct way.
Afterwards, you can use this lemma (and I'm sure there's a corresponding version somewhere in the standard library) to derive your additional hypothesis through apply.
Given any programming language, whenever a standard library function exists, we should most likely use it rather than write our own code. One would think that this advice applies equally to Coq. However, I recently forced myself to use the same_relation predicate of the Relation module, and I am left with the feeling of being worse off. So I must be missing something, hence my question. To illustrate what I mean let us consider to possible relations:
Require Import Relations. (* same_relation *)
Require Import Setoids.Setoid. (* seems to be needed for rewrite *)
Inductive rel1 {A:Type} : A -> A -> Prop :=
| rel1_refl : forall x:A, rel1 x x. (* for example *)
Inductive rel2 {A:Type} : A -> A -> Prop :=
| rel2_refl : forall x:A, rel2 x x. (* for example *)
The specific details of these relations do not matter here, as long as rel1 and rel2 are equivalent. Now, if I want to ignore the Coq library, I could simply state:
Lemma L1: forall (A:Type)(x y:A), rel1 x y <-> rel2 x y.
Proof.
(* some proof *)
Qed.
and if I want to follow my instinct and use the Coq library:
Lemma L2: forall (A:Type), same_relation A rel1 rel2.
Proof.
(* some proof *)
Qed.
In the simplest of cases, it seems that having proven lemma L1 or Lemma L2 is equally beneficial:
Lemma application1: forall (A:Type) (x y:A),
rel1 x y -> rel2 x y (* for example *)
Proof.
intros A x y H. apply L1 (* or L2 *) . exact H.
Qed.
Whether I decide to use apply L1 or apply L2 makes no difference...
However in practice, we are likely to be faced with a more complicated goal:
Lemma application2: forall (A:Type) (x y:A) (p:Prop),
p /\ rel1 x y -> p /\ rel2 x y.
Proof.
intros A x y p H. rewrite <- L1. exact H.
Qed.
My point here is that replacing rewrite <- L1 by rewrite <- L2 will fail. This is also true of the previous example, but at least I was able to use apply rather than rewrite. I cannot use apply in this case (unless I go through the trouble of splitting my goal). So it seems that I have lost the convenience of using rewrite, if I only have Lemma L2.
Using rewrite on results which are an equivalence (not just an equality) is very convenient. It seems that wrapping an equivalence into the predicate same_relation takes away this convenience. Was I right to follow my instinct and force myself to use same_relation? More generally, is it so true that if a construct is defined in the standard Coq library, I should use it, rather than define my own version of it?
You pose two questions, I try to answer separately:
Regarding your rewrite problem, this problem is natural as the definition of same_relation goes as double inclusion. I agree that maybe a definition using iff would be more convenient. It would really depend on the kind of goals you have. A possible solution for your problem is to define a view:
Lemma L1 {A:Type} {x y:A} : rel1 x y <-> rel2 x y.
Proof.
Admitted.
Lemma L2 {A:Type} : same_relation A rel1 rel2.
Proof.
Admitted.
Lemma U {T} {R1 R2 : relation T} :
same_relation _ R1 R2 -> forall x y, R1 x y <-> R2 x y.
Proof. now destruct 1; intros x y; split; auto. Qed.
Lemma application2 {A:Type} {x y:A} {p:Prop} :
p /\ rel1 x y -> p /\ rel2 x y.
Proof. now rewrite (U L2). Qed.
Note also that rewriting with a <-> relation is not really based on equality, but on "setoid rewriting". In fact, the following doesn't hold in Coq A <-> B -> A = B.
Regarding your second question, whether to use the Coq standard library is a highly subjective topic. I personally rarely use it, I prefer a different library called math-comp, but YMMV. Regarding relations, mathcomp is mostly specialized into boolean relations rel x y = x -> y -> bool, thus, equivalence is simply defined as equality, typically, given r1 r2 you'd write r1 =2 r2.
IMHO in the end, such choices are highly dependent on your application domain.
[edit]: Note that the Relation library is dated:
Naive set theory in Coq. Coq V6.1. This work was started in July 1993 by F. Prost.
So indeed, it may not be the best modern base to build Coq developments on.