The Notation of Harmonics for Bowed String Instruments
Author(s): Gerald Warfield
Source: Perspectives of New Music, Vol. 12, No. 1/2 (Autumn, 1973 - Summer, 1974), pp.
331-343
Published by: Perspectives of New Music
Stable URL: https://www.jstor.org/stable/832283
Accessed: 07-03-2019 17:48 UTC
REFERENCES
Linked references are available on JSTOR for this article:
https://www.jstor.org/stable/832283?seq=1&cid=pdf-reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms
Perspectives of New Music is collaborating with JSTOR to digitize, preserve and extend
access to Perspectives of New Music
This content downloaded from 132.204.9.239 on Thu, 07 Mar 2019 17:48:59 UTC
All use subject to https://about.jstor.org/terms
THE NOTATION OF HARMONICS
FOR BOWED STRING INSTRUMENTS
GERALD WARFIELD
The persistence of ambiguities and inconsistencies in co
musical scores with respect to string harmonic notation indi
the failure of various notational methods to receive gene
but also a widespread lack of understanding concerning the i
characteristics pertinent to the production of harmonics. Su
is probably the result of the relatively infrequent use of
the characteristic awkwardness of their fingerings (particular
ings for artificial harmonics). The following suggestions are
the hopes of providing a system that is unambiguous and th
adapt to projected notational changes. Further, it is hoped
information is given concerning characteristics of the instrum
the composer to use the suggested notation knowledgeably.
It should be mentioned at this point that although parts of
proposal will be familiar (and are already employed with var
of consistency) the second section may appear to be som
wards." That part, briefly, suggests that the tablature (stop
dicators) be placed within parentheses and that the soun
specified as a "normal" note. The advantages and disadvan
procedure are detailed later in the paper.
Proposed Notation
I. The sounding pitch should be indicated with a note of
and desired value to which is added a small circle just above
note head (or at a farther distance should the symbols becom
crowded together). This part of the suggested conventio
alone or in conjunction with either A or B of part II below.
II. Natural and artificial harmonics may be clearly distinguis
of the following:
* 331 '
This content downloaded from 132.204.9.239 on Thu, 07 Mar 2019 17:48:59 UTC
All use subject to https://about.jstor.org/terms
PERSPECTIVES OF NEW MUSIC
A. For natural harmonics the string should be indicate
with a Roman numeral, and the node should be in
low diamond note head in parentheses.1
x z
Ex. 1 Ex. 2. Natural Harmonics
B. For artificial harmonics the stopped
solid note head and the node with
head, both in parentheses.
Ex. 3. Artificial Harmonic
As stated above, one may employ only the first part of the suggested
notation omitting the tablature (part II) altogether; this grants the per-
former freedom to decide whether to play a natural or an artificial har-
monic, and, in some cases, choices of nodes and strings. (The performer
usually makes decisions of this kind on the basis of ease of fingering, habit,
and timbral differentiation.) Such a notational procedure is already em-
ployed by some composers, and, furthermore, as performers become in-
creasingly adept in the production of harmonics (which they will have to
become if they are attempting the contemporary literature) they will also
become less dependent on the tablature. Certainly if the notation of har-
monics undergoes any streamlining or simplification in the future it will be
the fingerings that will be deleted and not the sounding pitch. It would
thus seem reasonable to place in parentheses those indications that may
eventually become obsolete.
The timbres of natural and artificial harmonics are similar and at times
indistinguishable. If different, a natural harmonic is a little more open or
pure sounding than an artificial harmonic of the same pitch. This differ-
ence, however, is usually minimized as much as the performer's technique
will permit.
If the composer wishes to indicate that a note is to be played as a natural
harmonic but is uncertain of the criterion for the selection of nodes, it
would be best for him to omit the node specification. The Roman numeral
will still clearly indicate the type of harmonic to be produced.
1 Except in the case of harmonic arpeggios to be discussed later.
* 332
This content downloaded from 132.204.9.239 on Thu, 07 Mar 2019 17:48:59 UTC
All use subject to https://about.jstor.org/terms
NOTATION OF HARMONICS FOR STRING INSTRUMENTS
Many performers object to the omission of the nodes and
on the grounds that the physical gestures necessary to prod
cated pitches are not sufficiently represented. For instance,
much more effort than is immediately apparent while 4b re
Ex. 4a Ex. 4b
Switching between
perience with harm
ous needs for tabla
assume that as fami
the tablature part o
To elaborate on th
mentioned that for
join the stopped not
become confused w
clutter the parenthe
If several harmon
indications may be
Ex. 5
Furthermore, ties and slurs should join the notes indicating the sound-
ing pitch (Ex. 5a) and not the stopped note or node indicators (Ex. 5b).
)? R. @ ^t\. IqP _ti^ O .r
Ex. 5a Ex. 5b
The most serious cr
the tablature, lackin
tain complete perfor
if one reads the tab
fications.
Since the production
* 333
This content downloaded from 132.204.9.239 on Thu, 07 Mar 2019 17:48:59 UTC
All use subject to https://about.jstor.org/terms
PERSPECTIVES OF NEW MUSIC
or the thumb and a finger, every subsequent note usually
(assuming the same relative node with respect to the stopp
therefore, deceptively easy to notate configurations made
which are inordinately difficult to play (as shown in Ex. 6)
ple of a next-to-impossible configuration add a slur over th
Ex. 6
Advantages of the Suggested Convention
1. As harmonics are increasingly used in contemporary music, tablature
specifications (which are in addition to the pitch specifications) may be-
come less and less necessary. With the suggested convention one anticipates
this trend by placing in parentheses those parts of harmonic notation
which may eventually be discarded.
2. The degree of freedom given the performer is instantly recognizable,
and, if there is no freedom, the type of harmonic to be produced is clear:
a. If only the note and a small circle are given, it is the player's choice
as to the type of harmonic, string, and node.
b. If there is a Roman numeral it is a natural harmonic.
c. If there is a stopped note and a node in parentheses then it is an
artificial harmonic.
3. The symbols are used consistently (e.g., the diamond is always a node,
never the string or sounding pitch except in cases where the node is in a
corresponding position to the sounding pitch).
4. Specifying the sounding pitch as a note of normal size and value is con-
sistent with the principle of non-transposing scores.
Characteristics of the Instruments That Pertain to the Production
of Harmonics
On stringed instruments (as on brass) there is no clearly distinguishable
upper limit applicable to the range of possible harmonics. In general, the
higher the harmonic the less reliable it will be2 and the greater the noise
level. On the violin the fifth harmonic (sounding two octaves and a fifth
above the stopped note) seems to be the highest natural harmonic that can
be produced without special care or preparation.
2 This does not mean that the lowest harmonics are necessarily easy. On none of
the stringed instruments is the octave stretch, required for the production of the
first harmonic, possible in a low position with the possible exception of the violin.
See Characteristics of Harmonics on the Viola, Cello and Double Bass below.
* 334 -
This content downloaded from 132.204.9.239 on Thu, 07 Mar 2019 17:48:59 UTC
All use subject to https://about.jstor.org/terms
NOTATION OF HARMONICS FOR STRING INSTRUMENTS
Through a slurred harmonic arpeggio (sometimes referred t
monic glissando) even higher natural harmonics may be
sounding, but they are much less reliable. In such a case it is
harmonic-arpeggiation effect which is extended, and the prec
of each of the pitches should not be counted on. The violin ca
up to approximately the seventh harmonic. Such arpeggiation
effective method of preparation for the less reliable higher h
example, the fifth harmonic is easier to produce if slurred
fourth harmonic.
If one assumes that on the violin the fifth harmonic is the upper prac-
tical limit for natural harmonics, then there are only fourteen natural har-
monics on the instrument. Six of these can be played on two different
strings (at different nodes), and (by eliminating unison and octave dupli-
cations) there are a total of eight pitch classes represented in all (Ex. 7).
a m r I
io, iZOd, (,A)
Ex. 7. Natural harmonics available on
fifth harmonic as upper limit
Any natural harmonic except the fir
the desired pitch can be produced. Exa
diamond-shaped note heads) aligned
tion of a vibrating string for the first
violin.
Note, as in Ex. 8, that the relations
monic to one another from the bottom
harmonic series and that the top node
string as the sounding pitch (i.e., th
Since the second through the fifth ha
and there are four strings, sixteen no
with more than one node.
In the notation of natural harmonics
node nearest the nut is not always pra
Due to the high position necessary f
easier for the player to produce the h
same point (on the fingerboard) as the
shown above.
For the notation of artificial harmonics the node adjacent to the stopped
* 335 ?
This content downloaded from 132.204.9.239 on Thu, 07 Mar 2019 17:48:59 UTC
All use subject to https://about.jstor.org/terms
PERSPECTIVES OF NEW MUSIC
Sounding Pitch Nodes
1st Harmonic
2nd Harmonic
3rd Harmonic
MI t proJucet IL* hmnic
4th Harmonic 0
5th Harmonic , t
L^owces 1d wr_m
proAuces 2' har.mon
Ex. 8. Nodes of the first five harmonics
on the G string of the violin
Ai"f
Ex. 9
note is usually the only one specified. There are, however, two cases where
the second node can be reached (even in low positions) although not with
significant results. In the case of the fifth harmonic (produced with the
node a minor third away from the stopped note) the second node, a per-
fect fifth away, is also the first node for the second harmonic which will be
produced instead (Ex. 10). For the fourth harmonic the second node re-
quires the reach of a major sixth which is difficult and usually unnecessary
since the reach of a major third will produce the same effect (Ex. 11).
On the violin the lowest practical artificial harmonic is the pitch sound-
ing a twelfth above the lowest stopped note: the Eb a twelfth above the
* 336 '
This content downloaded from 132.204.9.239 on Thu, 07 Mar 2019 17:48:59 UTC
All use subject to https://about.jstor.org/terms
NOTATION OF HARMONICS FOR STRING INSTRUMENTS
P5
Ms3 3
bridge 4 ----* nut
t stopped note
first node
second node (but first n
of the second harmonic
Ex. 10. Fifth harmonic (p
'I M&
bridge 4 - O * - nut
t stopped note
first node
second node
Ex. 11. Fourth harmonic (produced artificially)
low Ab on the G string. This means that the range of all possible harm
ics on the violin starts on the second D above middle C and includes all
higher notes. The only other harmonic not included within this range is
the G an octave above the G string.3
Harmonic Arpeggio
The harmonic arpeggio is one case where the specification of nodes is
not recommended. The indication of the string and the term "gliss."
(which more accurately describes the action, not the sound) should be
sufficient as in Ex. 12.
,TZ. -I '
Ex. 12a Ex. 12b
Examples 15 and 16 s
arpeggios4 on specific
3 There are some violinis
aration time, can manage
thus extending the lower
C. See Paul Zukofsky's "O
174-81.
4 A natural harmonic played as a single note has an almost equal chance of
sounding at any of its nodes. Examples 15 and 16 show at which node a harmon
sounds in the course of a harmonic arpeggio.
* 337
This content downloaded from 132.204.9.239 on Thu, 07 Mar 2019 17:48:59 UTC
All use subject to https://about.jstor.org/terms
PERSPECTIVES OF NEW MUSIC
string player a clearer general picture of the mechan
production of natural harmonics.) Possible-or rather,
ics are shown with white note heads at actual pitch align
tive position of the nodes on a diagrammatical represent
Also shown is the position of the nodes with respect to
In the production of a harmonic arpeggio, the finger s
nodes which do not sound (as shown in Exx. 15 and 16
the viola C string the sounding of the harmonics one at a
in which their nodes appear from the middle of the strin
nut) would yield the pitches shown in Ex. 13.
Ex. 13
This would only occur, however, if each harmonic were bowed separately
or, perhaps, if the finger were moved very slowly. The usual slurred ha
monic arpeggio (i.e., a harmonic glissando) produces only the pitch
shown in Ex. 14.
Ex. 14
Although all of the nodes of a specific harmonic have a relatively equal
chance of sounding, it appears to be easier-in the context of a harmoni
arpeggio-for a string that is already producing a harmonic (as an appar
ent fundamental) to change to another harmonic (as an apparent funda
mental) when the new harmonic is an adjacency in the harmonic serie
(e.g., 1, 2, 3, 4, 5, 6, 7, 8 which is the series shown in Ex. 14 as opposed
to 1, 8, 6, 4, 7, 2, 6, 3, etc. which is the series shown in Ex. 13).
As a further result of this phenomenon the highest possible natural har-
monics on any string instrument can usually be produced only if they are
approached through a harmonic arpeggio (i.e., adjacent lower harmonics).
The sequential appearance of the nodes of these highest possible harmonics
at either end of the string makes it possible for the impetus or inertia of a
harmonic arpeggio to coax the string into vibrating in smaller and smaller
subdivisions thus ascending the harmonic series (see Exx. 15 and 16). As
mentioned earlier these higher notes are somewhat unreliable. In Exx. 15
and 16 they are represented with black note heads.
* 338 -
This content downloaded from 132.204.9.239 on Thu, 07 Mar 2019 17:48:59 UTC
All use subject to https://about.jstor.org/terms
-
-A Q a n
boWifg
PitcO
Nodc
Strin and
b1 . S .. 3$ ,>w1 t a r + (D-) o &- * '
I Node
lYF t * e c ss;+;on
^ (+) o (__*)
s l j /
7* t _ 7 ' ' , (7) 1 | t t7), ' (7)
G t - 6 ' $ , (G) l
' (G)
5 * _ 5 , § (5 l
(5)
4*+- 4 § $ (4)
3i+_ 3 | (3)
2 * _ 2
lft i _
1
tt All nodes of the first seven harmonics on the violin G string are shown. Those on the singl
o r of the string which are in parentheses will not wually sound in the context of a hanno
t .D
o
£ :F % If the node of more than one harmonic appears at the same point t
:i C>
Ex. 15
This content downloaded from 132.204.9.239 on Thu, 07 Mar 2019 17:48:59 UTC
All use subject to https://about.jstor.org/terms
Cello - D Strinq
So?sd)in, _| 'T- g.7 o - *fcf,
SP+ch, din i3 . ; I
11111 1 I I1
Node.., ", ,_e ' o
* 6Nodet G ~ a 4) 3
r o0' " (' (( /) | j) /
Po iII I I I
10 ^ _'5I
10' ' I
' , 'I(lo)l
I I I
(10) , , (,?'?5 I (5)
(10)1:'-" -!" 1
,t I
q'7-^ 6rs) < (7 ((7)
5",u t e, ,,!) (
2f I (4^) (7)
5 (5 ft Ahnodesof th firt .
eleen 2
hamonis on he celo (3)
Dstrin are hown.Thos on te (5
sinle-lne reresena
3 rd
'f "*"')t1
Iftenoeo mr ha n 'tin :tth ae o
hroicapar
All nodes of the first eleven harmonics on the cel o D string are shown. Those on the single-line representation
ta of the string which are in parentheses wil not usual y sound in the context of a harmonic arpeg io.
*=C gK~ '?If the node of more than one harmonic ap ears at the same point the lowest sounds as the fundamental.
Ex. 16
This content downloaded from 132.204.9.239 on Thu, 07 Mar 2019 17:48:59 UTC
All use subject to https://about.jstor.org/terms
NOTATION OF HARMONICS FOR STRING INSTRUMENTS
Since lower harmonics are easier to produce than higher one
tical to start on the node of a lower harmonic on or near the m
string and to slide either toward the bridge or toward the nu
the bridge is usually preferred).
Example 16 shows many of the nodes near the nut located a
points on the string. This is so because the physical distance b
jacent nodes of a harmonic is always the same no matter what
string they appear on, while the physical distance between eq
varies depending on whether the interval is in a high or lo
the string.
Harmonic Glissando
Harmonic glissandos must be produced as artificial harmonics because
of the gradual shifting of the fundamental. The straight gliss. line may
appear in the tablature parts as well as in the part specifying the actual
pitch, as shown in Ex. 17.
Ex. 17
Note that the physical distance (not the intervallic distance) from t
stopped note to the node will vary depending on the total length of
string segment that is actually vibrating. This means that for a harmon
glissando the hand cannot remain in a fixed position; if the glissand
ascending, the little finger must be brought gradually closer to the stopp
note to accommodate the shifting of the relative position of the node (j
as, for example, the physical distance of a perfect fourth is smaller
higher positions on a string).
The harmonic glissando should not be confused with the harmon
arpeggio (discussed earlier). Although both involve a continuous slid
motion, the sound of the harmonic glissando is an unbroken rise or fall
pitch while for the harmonic arpeggio the resulting sound is an arpeggi
tion of the pitches of the harmonic series of the particular string being
played (see Ex. 12).
Double Stops
Double stops where both notes are artificial harmonics are so difficult
produce and so closely related to physical limitations of the player and i
strument that they should not be notated unless the composer is certain
that they are possible within the context in which they are specified. D
ble stops made up of natural harmonics are usually easier and should
* 341 '
This content downloaded from 132.204.9.239 on Thu, 07 Mar 2019 17:48:59 UTC
All use subject to https://about.jstor.org/terms
PERSPECTIVES OF NEW MUSIC
clude two circles above the sounding pitches and the string
the same ascending order as the sounding pitches (and the n
gs
Ex. 18
Double stops which contain only one harmonic and one ordinary
stopped note range in difficulty from easy to impossible. It may sometimes
be helpful to reverse the directions of the stems (as in Ex. 19a) in order to
make it absolutely clear to which pitch the circle (harmonic) refers. In
Ex. 19a the parenthetical "harmonic" assures that the II is the string
specification for the harmonic and not the highest sounding pitch.
(hrrmeir) I
c?tto % ( f
(a) (b) (c)
Ex. 19
Characteristics of Harmonics on the Viola, Cello, and Double Bass
In the string family the larger the instrument the easier the higher ha
monics become due to the increasing ratio of the thickness of the strin
the size of the instrument and the width of the bow. However, it should
emphasized again that relatively higher harmonics are invariably m
difficult to produce and often contain greater degrees of noise element
than lower harmonics.
Viola. Practical range for natural harmonics can be extended to
sixth harmonic, although up to the eighth is possible through a harmon
arpeggiation. The artificial harmonic requiring the reach of a fourt
more difficult due to the larger size of the instrument and the artifici
harmonic requiring the stretch of a fifth is impractical except in h
positions.
Cello. The practical range for natural harmonics can be extended to the
seventh harmonic, although up to the eleventh is possible through a har-
monic arpeggiation. The sixth harmonic is very flat and may not be of use
in some contexts. Despite the increase in the size of the instrument, the
artificial harmonic requiring the stretch of a fourth is possible due to the
technique of playing the stopped note with the thumb. The harmonic re-
quiring the stretch of a fifth is more difficult.
* 342 ?
This content downloaded from 132.204.9.239 on Thu, 07 Mar 2019 17:48:59 UTC
All use subject to https://about.jstor.org/terms
NOTATION OF HARMONICS FOR STRING INSTRUMENTS
Double Bass. Artificial harmonics are impossible exce
positions. The thumb technique is used as on the cello. Nat
are very easy to produce and can be safely notated up to t
monic. Harmonic arpeggiations are possible to the ele
* 343 ?
This content downloaded from 132.204.9.239 on Thu, 07 Mar 2019 17:48:59 UTC
All use subject to https://about.jstor.org/terms
