Before we embark on an account of what could be called the liquidity preference theory of investment, a word of warning is needed. This is developed in Book IV, “The inducement to invest”, of the General Theory, which could be said to contain some of the most revolutionary and most original ideas of the entire book. That is certainly the view taken by Minsky (1975), who drew particular inspiration from chapter 17, as we are going to examine in a later chapter. This part of the General Theory is also the one that is in greatest need of recovery, however, in both the senses given to this term in Chapter 1. It needs to be recovered from its having just been reduced, in the neoclassical synthesis, to a respecification of a demand for money function that ends up providing an implausibly stark choice between investing everything in bonds and keeping everything as cash. But it also needs to be recovered from the potentially confusing terms in which Keynes chose to present it, thereby making its misrepresentation and eventual dismissal all too easy. As in other parts of the General Theory, there are crucial diagrams that really ought to have been drawn and not just described, and there are passages that seem to be understandable only by admitting that Keynes used mistaken or confused terms (Lerner 1952; Conard 1963). Most vexingly, Keynes was fully aware that he was the first to state explicitly assumptions of the orthodox theory and his own way of departing from them (see Moggridge 1973b), but this also means that the perceived obscurity of language and presentation can find very little hope of support and elucidation elsewhere. Scholars of Keynes have therefore found themselves confronted with the need to restate the terms of the theory in more lucid terms, but, in so doing, they have often been unable to capture all the various ideas that the chapters contain. Consistency has therefore proved to be elusive, and it will undoubtedly be so in what follows too. An eclectic choice has been made of what seem the most illuminating accounts, which are, essentially, Davidson (1972), Kregel (1982, 1988) and Conard (1963), with some effort at connecting them together.