Skip to content
Off-topic for aiqbench

Off-topic for aiqbench

This story was gathered from a source we follow but falls outside AI and quantum coverage, so there is no aiqbench take for it.

Hypotheticals

LessWrong4 min read

In contemporary philosophy, a hypothetical is a statement that considers what would, could, or will be the case under a specified condition. The term covers indicative conditionals as well as subjunctive conditionals, including counterfactuals. An ordinary hypothetical is a singular conditional statement ("if p, then q") concerning a particular (i.e. singular, dated or datable) case. Ordinary hypotheticals resemble the material conditionals studied in formal logic because they have the same surface form ("if p, then q"). The Law of Excluded Middle states that every proposition is either true or false. Any genuine proposition satisfies the Law of Excluded Middle on its own terms: taken as a whole, it must be either true or false. A material implication is a proposition that is either true or false according to the truth values of its antecedent and consequent. Frank Ramsey argues that the role of ordinary hypotheticals in reasoning is different than the material implication. An ordinary hypothetical functions as a conditional judgment whose evaluation depends on supposing the antecedent rather than assigning a truth value to the conditional itself. Take the ordinary hypothetical ‘If I were you, I would take the bet’ The antecedent ‘If I were you …’ is false, yet under the material implication, the conditional ‘If I were you, I would take the bet’ is (vacuous) true. In practice, an agent does not assess whether ‘If I were you, I would take the bet’ is (vacuous). Instead, an agent assesses whether they agree or disagree with ‘If I were you, I would take the bet’.  An ordinary hypothetical as a whole is never assigned a truth-value. Only its consequent is, and only conditionally, once the antecedent is supposed true. Since the whole sentence ‘If I were you, I would take the bet’ is not simply true or false, it fails the test that marks a genuine proposition. Frank Ramsey concludes ordinary hypotheticals are not genuine propositions, but sentences that express cognitive attitudes. An agent does not treat ‘If I were you, I would take the bet’ as a proposition that can be true or false, but as a conditional judgement that can be agreed or disagreed with. To understand how an agent makes a judgment to agree or disagree with an ordinary hypothetical, it is useful to classify ordinary hypotheticals based on two dimensions: time and modal status: No. Hypothetical type Time Modal status 1 Present open hypothetical Present Open 2 Past open hypothetical Past Open 3 Future open hypothetical Future Open 4 Present unreal (counterfactual) hypothetical Present Unreal (counterfactual) 5 Past unreal (counterfactual) hypothetical Past Unreal (counterfactual) 6 Future unreal (remote) hypothetical Future Unreal (remote) Time Time refers to when the event named in the antecedent (the "if" clause) is supposed to occur, relative to the moment of speaking. There are three possible values. 1. Present means the event is happening now, or is understood as a current state of affairs. 2. Past means the event has already happened, or is claimed to have already happened, before the moment of speaking. 3. Future means the event has not yet happened and is placed after the moment of speaking. Time does not say anything about whether the agent believes the event actually occurred or will occur. It only fixes where the event sits on the timeline.  Modal Status Modal status refers to whether the agent treats the antecedent as open or as false. It does not say anything about when the event occurs; that is the role of Time. Modal status has two possible values: Open or Unreal. 1. Open means the agent does not yet know, or does not yet treat as settled, whether the antecedent is true or false. The matter remains undecided from the agent's point of view. 2. Unreal means the agent treats the antecedent as false, or as going against what is actually the case. An agent cannot make a judgement to agree or disagree with an ordinary hypothetical in isolation. To make that judgement, an agent needs to hold a ‘rule for judging’. Ramsey refers to these rules for judging as variable hypotheticals. An agent asserting ‘If I were you, I would take the bet’ might hold the rule for judging/variable hypothetical:  "If I am offered a bet with positive expected value, at stakes I can afford given my risk tolerance, I will take it." This specific variable hypothetical forms singular beliefs that link ‘a bet with positive expected value, at affordable stakes given the agent’s risk tolerance’ to ‘accepting a bet’ with an affirmative character across future instances. When the antecedent of the variable hypothetical is satisfied by a particular case, it produces a singular belief in which the linkage is made and asserted for that case. Two agents who hold different variable hypotheticals may therefore agree or disagree differently with the same ordinary hypothetical, without either being mistaken about the singular case itself; the

Read the full article at LessWrong

Related coverage