Ineffability is a traditional topic of philosophical and religious discourse, discussed by Plotinus, Meister Eckhart, John of the Cross and Al-Ghazali, among others (see Alston 1975 and Hick 2000). More recently, it has been considered in relation to idealism, scepticism and practical knowledge (Hofweber 2016, Jonas 2017, Kukla 2006, Moore 1997). The most famous discussion of ineffability appears in Ludwig Wittgenstein’s Tractatus Logico-Philosophicus, where one finds the distinction between “things” that can be said and “things” that cannot be said, but only shown. Surprisingly, it is difficult to find an exhaustive analysis of “things” that cannot be said but only shown in the classical Wittgensteinian literature (Anscombe 1963, Geach 1975, Hacker 1986, Malcolm 1986, McGuiness 2002, Pears 1987). In the first part of this talk, I characterise what it means that something cannot be said, but only shown. I argue that Wittgenstein distinguished four types of ineffability (1. indescribability; 2. unassertability; 3. uninformativeness; and 4. ineffability per se). Each type of ineffability cannot be said for a different reason, and each has different philosophical consequences for what we cannot know and for the kinds of things we cannot know. In the second part of the paper, I analyse in detail which kinds of content are ineffable and what sorts of knowledge about them are unattainable according to the Tractarian conception.