Aural Comprehension Guide

David Loberg Code          Western Michigan University

Alphabet Dictation

If you are someone that cannot remember anything if you wait until after the melody is played to begin writing, or cannot get past the first measure if you don't wait and begin writing immediately, you may find this technique helpful. Write out the letters of the alphabet (with plenty of space) ahead of time as a framework. Each note that you hear will correspond to a letter. For example, if the rhythm contains nine notes, you should reach the letter 'I', the ninth letter of the alphabet.

This way, you can 'follow along' (with your eye and/or finger) even on the first hearing, and you can fill in the details of the melody or rhythm gradually.
In the beginning, I recommend that you focus on only one musical feature at a time, separating melodic and rhythmic elements as well. Naturally, this necessitates considerably more hearings than normally provided for dictation exercises, but after a little practice you will be able to combine steps together. The following are the initial sequences I suggest for melodic and rhythmic materials:

1. Length: On what letter does it end?
2. Contour: Is each note above, below, or the same as the one before it?
3. Conjuct vs. Disjunct: Is each note a step (minor or major 2nd) or a leap (minor 3rd or larger) from the one before it?
4. Patterns: Do you hear any scalar segments, arpeggiation of triads, prominent scale degrees, non-consecutive repeated notes, altered tones?
5. Specifics: Identifying specific pitches or intervals.

1. Length: On what letter does it end?
2. Measures: On or between which letters does each downbeat occur?
3. Beats: On or between which letters does each interior beat occur?
4. Equal vs. Unequal Divisions: Do the notes divide each beat into equal or unequal parts?
5. Patterns: What is the general pattern (rhythmic contour) for each unequal division (e.g., long-short, or short-short-long, etc.)
6. Specifics: Identify specific durations and rhythms.

Rather than working left-to-right, completing each measure before moving on to the next, this layered approach begins with very general information about the whole melody and gradually fills in successively finer details. Separating the listening process into distinct layers like this also helps identify areas of weakness. For example, if you cannot easily identify the direction of intervals (up or down) it is not surprising that you get the notes wrong, and it is of little value for you to continue to subsequent levels. The same applies when difficulties are encountered at any of the other levels. You should stop and spend your practice time mastering the more basic skills before progressing further.

Using the alphabet framework, the details are notated in a fairly simple shorthand that is translated into traditional notation at a later stage of the dictation process. Below is an example of an alphabet dictation of a melody:

Original Melody

Alphabet Dictation

The length of the excerpt is indicated by a double-bar following the the letter corresponding to the last note. Barlines and beat divisions (half barlines) are placed between the appropriate letters. Contour is notated by a dot above or below the space in-between consecutive letters (or within the space itself for unisons). Stepwise motion is notated by a dash (hyphen) between letters; leaps by a blank space. In places where you cannot decide whether the interval is a step or a leap, put a question mark. This not only prevents mistaking and unfilled space for the deliberate blank 'notated' to indicate a leap, but also highlights the specific intervals for whichyou should listen during the next hearing. Arpeggiation of triads are indicated by triangles; scalar passages by inclusive square brackets above the letters; non-consecutive repeated notes by connecting slurs; and non-consecutive stepwise motion by angled lines (above or below the letters depending on the direction of the interval). Although not present in the example above, chromatic tones are indicated by asterisks. Additional melodic details such as intervals, scale-degrees, solfege syllables, or pitch names are added in as appropriate. Rhythmically, an equal sign or a not-equal sign indicates whether a beat is divided into equal or unequal parts. In the case of the latter, the general rhythmic contour is specified by relative durations: L = long, M = medium, S = short. An important aspect of this shorthand notation is that it can be used in real-time while listening to the musical excerpt (which is not possible when writing noteheads, stems, and flags). Hence, those of you who have difficulty remembering what you heard after the excerpt is over, can notate information while you hear it without getting behind the music.

Initially, alphabet dictation is more time-consuming (but perhaps more accurate) than a traditional approach, especially if you must concentrate on only one piece of information at a time. After a little practice, however, you should be able to group together several features at once, and thus alphabet dictation can employ the same number of hearings used in regular dictation. For example, the first time through you may

1. mark-in on which letters each down-beat occurs
2. put a mark above or below each letter indicating whether the melody goes up or down
3. mark each letter that corresponds to the tonic (Do)

On the next hearing, you could refine this information as follows:

1. mark-in each beat within a measure
2. indicate whether each melodic interval is a step or leap
3. connect repeated notes
4. indicate general patterns (triads, scales)

In this way, you are starting from general skeleton of the whole melody, and gradually filling in more details. You won't always need all of the details in order to transcribe the melody. By logically piecing together the various bits of information (i.e. applying what you've learned in theory class) you can reconstruct the 'missing' notes and complete the melody. The first step is to combine all of the information already notated in the alphabet dictation. Returning to the alphabet dictation shown above, if the last note 'V' is Do, and the preceding note 'U' is a diatonic step above that, than 'U' must be Re. Similarly, 'T' is a diatonic step above 'S', which in turn is a diatonic step above 'U' making the entire last measure Mi-Fa-Re-Do. Using similar deductive methods it is possible to reconstruct all of the notes from the above example, except 'C' and 'Q'. Regarding the latter, given that it is part of an arpeggiation of a triad that starts on Do and ends on Mi, there are only two options: Do-La-Mi (a vi chord) or Do-So-Mi (a I chord). If there is an additional hearing available, you could listen for which one of these two pitches occurs. Even if there were no more hearings available, the odds of guessing the correctly pretty good with only two possibilities.
A further means of deducing or intelligently guessing missing information is to apply ones knowledge of the style and/or parameters in which the excerpt was composed. You have already learned the melodic and harmonic conventions which underly tonal melodies in theory class. These are the same kinds of melodies. For instance, by determining the implied harmonic progressions and cadences of a given melody, you could make a reasonable guess about a particular pitch or set of pitches. In measure one above, you might guess the missing note at letter 'C' to be Re because it would outline a V7 chord in beat one resolving to an implied I chord in beat two. Besides following general stylistic conventions, the musical materials employed in ear training texts and courses are usually introduced in graduated levels with well-defined boundaries. You should take advantage of these graded levels to guide your listening and decision-making strategies. Let us suppose that the level of rhythmic complexity in Example 2 were limited to the second division of the beat (i.e., sixteenth notes) without the use of syncopation or borrowed divisions. In that case, the information provided regarding equal and unequal divisions of the beat (along with the general rhythmic contour) would be sufficient to determine the duration of each note. Moreover, this approach encourages listening for recognizeable patterns, rather than trying to aurally measure the length of each duration separately as it is heard. Melodically, we may be working only with leaps from the tonic, subdominant, and dominant-seventh chords and thus the note 'Q' could not be La since it would then outline a submediant chord.

Alphabet dictation can be very useful for some people. Initially, it may be more time-consuming (but more accurate) than a more traditional approach. At first, you may need to concentrate on only one piece of information at a time (e.g., only listening for leaps or steps with nothing else). With practice, however, you will be able to group together several features at once. Most of you will probably not rely exclusively on alphabet dictation, but will add these new skills and ways of thinking to your existing repertoire of techniques.


David Loberg Code, School of Music, Western Michigan University, Kalamazoo, MI, 49008. E-mail: 
Revised: 28.Feb.99       (c) 1999