Closely related (sub)question: is there some way to tell a piece of (spoken, or writen, or...) text that is an argument from one that isn't? If 'yes', how?
[Notice the question is not asking "What is a valid argument? Is there a way to tell a valid argument from an invalid one?", as this assumes some notion/definition of 'argument' is already in place]
Yes, there's an entire field of study called argumentation theory which is essentially the philosophy of arguments. There are different models, in Uses of Argument (GB), Stephen Toulmin lays out a good model of argumentation now named after him. From the WP article on argumentation theory:
Argumentation theory is the interdisciplinary study of how conclusions can be supported or undermined by premises through logical reasoning. With historical origins in logic, dialectic, and rhetoric, argumentation theory includes the arts and sciences of civil debate, dialogue, conversation, and persuasion. It studies rules of inference, logic, and procedural rules in both artificial and real-world settings.
Generally, arguments in reason are classified as deductive, inductive, or abductive, or some mix thereof, and the broader notion of argumentation which might use such persuasive factors as emotions, testimony, and fallacies is the object of study under rhetoric which views argumentation as a process that is liable to be highly irrational.
Without getting technical, you have a argument when a conclusion is said to follow from a set of premises. For a deductive argument, If it in fact follows, the argument is classified as valid. Otherwise the argument is invalid. Inductive arguments are neither valid nor invalid.
To your subquestion- yes, there is a way to tell you have an argument. The conclusion will be preceded by words like, Therefore, So, Thus, Hence, etc.
If you want to get technical, you have an argument when a corresponding conditional is claimed to be a tautology.
Suppose you have a set of propositions denoted by statements A1, A2, A3,... An, which we will call premises, and a proposition denoted by B, which we will call the conclusion, if someone claims the proposition denoted by the statement
"IF A1 & A2 & A3 &... &An THEN B"
is a tautology, then they are claiming the conclusion follows from the premises. Thus they formed an argument.
Epistemic virtue allows us to identify reasons of an argument, which only goes beyond its premises if it is defeasible.
Reasoning is defeasible when the corresponding argument is rationally compelling but not deductively valid.
I would think that to insist on non-defeasible arguments would only have the appearance of epistemic virtue.
