This article provides a new interpretation of the linguistic aspects of Friedrich von Hardenberg’s Fichte Studies. It argues that Hardenberg was searching, among other things, for a transcendental language for philosophy. The possibility of such a language was discussed intensely among his contemporaries, such as Maimon, Niethammer, Reinhold, Weißhuhn, and Fichte. Its necessity, however, had become apparent with Kant’s Critique of Pure Reason. Readers had noticed a disturbing discrepancy between the objective knowledge of transcendental philosophy—which, according to Kant, was supposed to be generally communicable—and Kant’s actual failure to communicate it. Hardenberg’s original insight into the inseparable unity of sign and signified, anticipating modern linguistic theories, led him to the assumption of a lawful relationship between both. From his unsuccessful attempt to disclose these laws, he went on to discover language as an independent realm fundamentally opposed to nature. Precisely because language is a necessary illusion, only the ‘presenting I’ (das darstellende Ich) achieves its end, namely absolute freedom. Philosophy, therefore, is pure as long as it remains within the boundaries of language alone, that is a language which does not refer to anything outside itself.