The notations used for chant from the ninth to the early eleventh century give very little information about pitch. As I discuss in Chapter 2, all the early dialects show melodic direction and the littera significatiuae indicates where two consecutive notes lie at a unison. But for the most part, pitch relations remain opaque. The fact that musicians knew three systems for the accurate presentation of pitch in the tenth and eleventh centuries places this situation in high relief: the palaeofrankish and dasian notations mentioned in Chapter 2, to which we can add the alphabetic notation advocated by Hucbald of Saint Amand derived from Boethius and that used in Norman manuscripts.

By the early twenty-first century, Western musical notation has developed into an extraordinarily rich and complex piece of visual communication. Modern musicians expend considerable time and energy to create strategies for its more efficient comprehension. Simultaneously, musicians are continuously inventing new symbols and systems of symbols for two very different reasons. Some wish to communicate with ever greater precision the instructions necessary for the reproduction of a musical work. Others are seeking ways of recording in writing musics whose origins are not necessarily rooted in notation. Charles Seeger identified these two streams as, respectively, “prescriptive” and “descriptive” forms of notation.

From its inception in the medieval West, musical notation constituted a complex system of visual communication that required extensive and specialized knowledge of the musical practices it attempted to depict along with a fine visual discrimination and apprehension of the symbols, to translate them from symbol to sound. Consequently, musical literacy has remained, since its introduction, the exclusive product of professional training at a high level, and, in some areas of musical practice, an indispensable credential for achieving professional status. Yet, this system simultaneously exerts enabling and limiting forces on musical practice. Obviously, a musician separated by distance and time from the creation of a musical work is able to read, study and perform it through the efficacy of notation and literacy. The ability to execute these tasks depends on the musician’s competency in musical literacy as well as her or his knowledge of the notation and musical style of the work, strictly historical considerations.

Beginning around A.D. 1000, musicians cultivated increasingly complex styles of polyphony. The earliest practical polyphony, from Winchester, Aquitaine and Paris, simply employed the prevailing style of plainsong notation. By the middle of the twelfth century, that notation accurately specified the pitch content of a melody, as I discuss in Chapter 2 and Interlude 1, and so the pitch content of the simultaneously sounding voices became secure. Rhythmic coordination in the Winchester polyphony of the eleventh century caused no problems because it proceeds in strict note-against-note style, and so each note in the organal voice corresponds to a single note in the liturgical chant. Aquitanian polyphony of the next century, however, often sets more notes in the upper voice than in the lower, and so debates ensue as to how to orient the two voices.

The earliest written instrumental music uses contemporary vocal notation: the dances in Pa 844 (thirteenth century), Lo 28550 (the Robertsbridge codex, fourteenth century), Lo 29987 and Faenza 117 (both fifteenth century). The Robertsbridge codex contains a hybrid notation in which the upper voice or voices appear in conventional vocal notation and the lower voice is represented by the letter names of the notes. This system becomes the ancestor of German keyboard tablatures of the fifteenth century, which similarly use conventional and literal notation for the upper and lower voices, respectively. These developments signal independent and idiosyncratic approaches to notation that become specific for each instrument or group of instruments, which we today collect under the generic term tablature.

The tension between surface rhythm and the organization of musical time into a regular succession of pulses existed from the beginning of rhythmic notation in the West. Johannes de Garlandia, writing about the rhythmic modes in the mid-thirteenth century, cites a rule (“Unde regula”) that “every note that falls in an odd-numbered position ought to concord with each odd-numbered note” (“omne, quod fit impari, debet concordari omni illi, quod fit in impari”). This stipulation falls well short of constituting a comprehensive theory of musical metre, but it does seem to discourage us from perceiving rhythmic music of this period as consisting of a series of undifferentiated beats. Still, thirteenth-century notation makes no provision for indicating where stress should fall. The transition from modal to mensural notation in the second half of the century did not bring any further clarity to the situation.

Western art music distinguishes itself from the music of many other cultures by the use of polyphony. The need to show the simultaneous progression of two or more lines of music has generated a number of novel solutions over the millennium or so that Western musicians have practised polyphonic music. Monophony, of course, poses no particular problems of layout. Medieval scribes of monophonic music chose to write the music from left to right, borrowing the conventions from writing text, largely because monophony was texted. They inscribed the music either all the way across the page or in columns, but that choice did not materially affect the presentation of the music or its apprehension in reading or copying. The earliest practical sources for polyphony, the Winchester tropers of the early eleventh century, write the voices separately as monophony, in distinct sections of the book or even in two books.

The continuous history of musical notation in the West begins in the ninth century. Two groups of competing theories prevent more definitive statements about those origins, and their critical treatment serves as a starting point for the narrative. One group of theories concerns the origin of the neumes themselves, of which the most recent scholarship recognizes three possible sources: punctuation signs in Latin, Latin accentual signs and a form of graphic notation that was eventually replaced by one that represented the melodic contours of the chant. The other concerns the date at which notation originated in the Latin West, and the two competing theories present the opposite ends of the ninth century for its beginning.

Over the last century, composers have exercised increasing control over every aspect of performance. They have adopted several strategies, including Stravinsky’s program of recording performances of every piece he composed with himself as pianist or conductor in order to provide templates for all other performances. The principal means for achieving this control, however, lay in the density and specificity of notational instructions. Composers were already using notation to assume control over aspects of performance before the end of the sixteenth century, perhaps beginning with the dynamic marks Giovanni Gabrieli introduced in his Sacrae symphoniae of 1597. A significant increase in the specificity of notation occurs in the works of Beethoven, who, because of the physical infirmity of his deafness, was the first major composer unable to participate effectively in performances of his own compositions.