Perspectives

Studies in Translatology

ISSN: 0907-676X (Print) 1747-6623 (Online) Journal homepage: http://tandfonline.com/loi/rmps20

Dubbing of silences in Hayao Miyazaki’s Spirited

Away: A comparison of Japanese and English
language versions

Reito Adachi

To cite this article: Reito Adachi (2016) Dubbing of silences in Hayao Miyazaki’s Spirited Away:
A comparison of Japanese and English language versions, Perspectives, 24:1, 142-156, DOI:
10.1080/0907676X.2015.1024694

To link to this article: http://dx.doi.org/10.1080/0907676X.2015.1024694

Published online: 28 Aug 2015.

Submit your article to this journal

Article views: 107

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

http://tandfonline.com/action/journalInformation?journalCode=rmps20

Download by: [University of Tehran] Date: 17 February 2016, At: 05:50

 Perspectives: Studies in Translatology, 2016
Vol. 24, No. 1, 142–156, http://dx.doi.org/10.1080/0907676X.2015.1024694

Dubbing of silences in Hayao Miyazaki’s Spirited Away: A comparison

of Japanese and English language versions
Reito Adachi*

Department of Early Childhood Education and Care (English Section), Kurashiki City College,
Kurashiki-shi, Japan
(Received 22 March 2014; accepted 17 February 2015)
Downloaded by [University of Tehran] at 05:50 17 February 2016

This paper clarifies how silences in a particular Japanese animated film are dubbed in the
US English version. As a representative sample of Japanese animated films, the author
examines the silences in Hayao Miyazaki’s Oscar-winning animated fantasy Sen to
Chihiro no Kamikakushi (Spirited Away), focusing on its acoustic modifications. First,
silences in the Japanese and US English versions are identified using Adobe Audition®
CS6. Next, silences in the two versions (acting as specific case studies) are compared
with each other to examine the process involved in dubbing them. A quantitative
comparison of the number of silences between the US English version and five other
foreign-language versions (the versions released in Taiwan [Republic of China], France,
South Korea, the Czech Republic, and Germany) is then carried out. In conclusion, the
US English version of Spirited Away removes more silences than any other translated
version by inserting fillers and by adding or amplifying sound effects. These findings
indicate the importance of studying audiovisual translation (AVT) not only from the
verbal perspective but also from the acoustic perspective.
Keywords: dubbing; silence; Hayao Miyazaki; Spirited Away; audiovisual transla-
tion (AVT)

1. Introduction
This paper examines silences in Hayao Miyazaki’s Oscar-winning film Sen to Chihiro no
Kamikakushi (Spirited Away),1 which was released in the United States in 2003. In
audiovisual translation (AVT), it is necessary to deal with multimodal information, which is
broadly classified into screen images and sounds (Delabastita, 1990), so as to maintain a
balance between adequacy and acceptability of the ‘whole situation’ (Oittinen, 1993, p. 85).
Professional dubbing projects follow a complex and multifaceted process. The needs,
demands, and expectations associated with each step are fulfilled by individuals with
various skills, including translators, adapters (dialogue writers), synchronizers, dubbing
directors, producers, voice actors, dubbing companies (automatic dialogue replacement
productions), distributors, and the producers of the original version (Chaume, 2012, pp. 29–
39; Martinez, 2004). During this long, painstaking process, according to Martinez (2004),
they ‘may be more or less respectful of the original translation’ (p. 3) written by translators.2
This paper is part of an extensive study investigating various aspects of English translations

*Email: [email protected]
This work is an expanded and revised version of the paper presented at the International Conference
on Language, Literature, and Linguistics, 2013, in Colombo, Sri Lanka, on 2 December 2013.

© 2016 Taylor & Francis

 Perspectives: Studies in Translatology 143

of Japanese animated films: how they are translated, how they are received in the target
language, what parts are translatable, and what parts defy translation.
The following research questions are posed in this study.

(1) How does the US English translation of Spirited Away localize silences?
(2) Is the US English version a global standard for translating silences?

From the acoustic perspective, the author examines the research on the effects of silences
in the US English translation of Spirited Away. Because it is beyond the scope of this
paper to discuss Miyazaki’s art and craft in its entirety, the author leaves topics such as
Miyazaki’s characterization, stories, worldview, and graphics for other studies.

2. Silences
Downloaded by [University of Tehran] at 05:50 17 February 2016

The word ‘silence’ has many shades of meaning. For example, in his detailed discussion
of silence, Poyatos (2002) describes silence ‘as a component of interaction’ (p. 299).
Silence commonly means the complete absence of sound. The word can also refer to a
period of time when no one speaks. Silence may also imply the refusal to talk about
something or to provide information. In this paper, however, silence refers to the presence
of sounds of very low intensity or the lack of audible sound for a certain period of time.
To be more exact, silence is defined in this paper as a period during which sounds are
lower than −40 decibels relative to full scale (dBFS) for longer than 10 seconds. To
measure the sound level of each audio file, the author used the sound editing software
Adobe Audition®. In Adobe Audition®, the maximum possible amplitude is 0 dBFS; all
lower amplitudes are expressed as negative numbers. The sound intensity level of 0 dBFS
is the maximum amplitude possible; that of –20 dBFS, for example, is the reference level
to which broadcast engineers in North America usually adjust their audio equipment
(a status known as ‘broadcast safe’). The loudness level of spoken dialogue in a movie is
required to be a minimum of –31 dBFS according to the dialnorm parameter, which is an
indication of the average loudness of normal speech within an audio program (Williams,
Jones, Layer, & Osenkowsky, 2007, p. 1324).3 A sound of –40 dBFS is equivalent to
barely audible background music or really quiet sound effects such as a faint rustling of
leaves.
There are three main reasons why the author chose to examine the topic of silences.
The first reason for choosing this topic is that the frequent use of silences is a marked
characteristic of Japanese films compared with US films, reflecting the differences in
conversational styles between Japanese and Americans. For example, H. Yamada (1997)
reported that the average rate of silence during a Japanese business meeting is 5.15
seconds per minute but is only 0.74 seconds per minute during a US business meeting. To
explain this difference, H. Yamada suggested that ‘for the actual Japanese commu-
nicators, silences demand the active listening work of sasshi [guesswork], and it is
anything but hollow’ (Yamada 1997, p. 77). Miyazaki, acknowledging that a pause of
silence is a distinctive feature of Japanese discourse and that it affects the flow of
dialogue in his films, offered the following explanation: ‘We have a word for that in
Japanese. It’s called ma. Emptiness. It’s there intentionally’ (Ebert, 2002).
The second reason for choosing this topic is the effectiveness of silences in animated
cartoons. The 2009 Oscar-winning animated Japanese short film Tsumiki no Ie (The
House of Small Cubes) offers some examples of such effectiveness. The film uses the
inherent visual characteristics of animated cartoons, such as elastic deformations, shape
 144 R. Adachi

transformations, hyperkinetic movements, and color changes, which are powerful enough
to allow substantial freedom in the expression of emotions and the telling of stories
without having to depend on dialogue. There are many other animated films without
spoken dialogue; for example, the silent images portrayed on-screen in WALL·E are
often as eloquent as words, especially during the first 40 minutes of the movie,
which contain not one line of dialogue. Animated films, more so than live-action movies,
have a tendency to tell their stories mainly through image and sound (Adachi, 2012b,
pp. 43–44).
The third reason for choosing this topic is that although silences are usually outside
the scope of translation studies of texts, they may be analyzed in the studies of AVT.
Thus, the examination of this subject may be crucial to understanding the translation of
Japanese animation in general.
Downloaded by [University of Tehran] at 05:50 17 February 2016

3. Previous studies
3.1. Acoustic approaches to AVT
Poyatos, who exhaustively examined the theoretical and empirical issues associated with
nonverbal communication, described silences in narrative texts (1997a, pp. 35–37) and
interpretation situations (1997b, pp. 254–255, 266–267). He identified six basic functions
of silences: ‘signs proper’, ‘zero signs’, ‘carriers of the preceding activity’, ‘anticipator of
the succeeding activities’, ‘receptacles for light, sound and other signs’, and ‘enhancing
framing for the visual environment’ (2002, p. 303). In his comprehensive and accurate
overview of AVT, with a focus on dubbing in AVT, Chaume (2012) observed that
translators, dialogue writers, and dubbing assistants in some European countries usually
type single or double slashes (forward or backward) in translation to indicate the places
where pauses and silences are located in the original dialogue (pp. 58–63). Among those
who analyze acoustic elements in AVT, Palencia Villa (2002) pointed out that fictional
characters in a Spanish dubbed film are as credible to Spanish audiences as characters in
its original English version, demonstrating that the viewers’ perception of the dubbed
characters depends largely on how their acoustic elements, including voice quality, are in
accordance with the viewers’ ‘wish to enjoy the diegetic experience and the genre effect’
(p. 68). Building on Palencia Villa’s investigation, Bosseaux (2008) focused on a song in
a French dubbed version of a musical episode of an American TV series from the
viewpoint of multimodal parameters for character construction. She showed empirically
that shifts in aural elements, such as the pitch or tempo of an actor’s voice, affect viewers’
perceptions of the characterization; that is, the interplay between linguistic and
nonlinguistic cognition.

3.2. Approaches to the translation of Miyazaki’s films

The translated versions of Miyazaki’s animated films have so far been discussed mainly
from a linguistic perspective in the literature. While conducting a comparative analysis of
the original version of a Miyazaki film and its translation, Elwood (2003) investigated the
differences in the language used by characters in expressing requests. Gonzalez (2009)
examined how cultural elements in Miyazaki’s films are translated into English by
conducting a descriptive and comparative analysis of the films’ subtitles. However, little
attention has been given to the acoustic aspect of translated Japanese animation, with rare
exceptions, such as Teshigawara’s (2003) investigation of the phonetic correlates of vocal
 Perspectives: Studies in Translatology 145

stereotypes through analyzing the phonetic properties of the voices of heroes and villains
in anime.

3.3. Previous findings about silences in Miyazaki’s films

The translation of Japanese animated works underwent varied degrees of modification
during the 1980s and 1990s (Kusanagi, 2003; Okuhara, 2009). The first English version
of Miyazaki’s Kaze no Tani no Naushika is a notable example. The animated movie,
which was released in the United States as Warriors of the Wind in 1986, is infamously
renowned for undergoing drastic rewriting and severe editing before its release. The
American version cuts 21 minutes from the original film and deletes more than 30% of
the sentences in the source version to make the length more child friendly. The names of
most of the main characters in Kaze no Tani no Naushika are substituted with ones that
Downloaded by [University of Tehran] at 05:50 17 February 2016

are more English-sounding. The main storyline is also simplified to that of a war film of
didactic moralism (Adachi, 2012b, pp. 142–147).
Yet, as Japanese animated works achieved greater recognition in the United States
during the first decade of the twenty-first century, their translations have subsequently
become more faithful to the original versions. This change was accelerated by the
growing popularity of fansubbing and fandubbing, which were initially intended to allow
viewers to enjoy the unaltered forms of original works. These new practices have been
discussed from various angles, including legality. For example, in his historical analysis
of fansubs, Leonard (2004) suggested that the act of fansubbing spurred the growth of
Japanese anime in the North American market. In addition, Díaz Cintas and Muñoz
Sánchez (2006) highlighted how the practice of posting amateur translations of Japanese
anime in cyberspace opened the world of translation to new possibilities for creativity.
Pérez-González (2006) also pointed out the importance of studying the viewer-centered
movement of fansubbing, which has made a substantial contribution to the globalization
of Japanese animated movies. Although problems of copyright infringement and product
quality (Bogucki, 2009) are associated with fansubbing and fandubbing practices, it may
be safely assumed that they have also acted as advertisements for the licensed versions of
Miyazaki’s movies and even positively affected the animation market and industry as a
whole.
Although the general tendency is toward a more faithful approach, the avoidance of
silences continues today in translations. Silences, as well as taboo topics such as
sexuality, drugs, and violence, in Miyazaki’s movies are part of the culture-bound
category that often defies literal translation (Adachi, 2012a). This phenomenon is
illustrated by Adachi (2010), who used a linguistic approach to examine 11 Japanese
animated movies and 11 US animated movies.4 This research led to three important
discoveries.

(1) On average, Japanese animated films have more silences per 10 minutes (7.56
silences) than US animated films (4.15 silences per 10 minutes).
(2) The number of silences in Japanese animated movies decreases significantly in
the process of English translation. However, the number of silences in US
animated movies does not decrease when translated into Japanese. It is clear that
US culture has a much stronger influence on the translation of Japanese
animated films into English, but as far as US works are concerned, Japanese
translation is relatively faithful to the original English-language films. Hence,
 146 R. Adachi

there seems to be a marked imbalance in the translational attitudes about silences

between the two languages.
(3) Despite the general trend toward a more faithful approach in the translation of
Japanese animated movies, the practice of avoiding silences persists today.
Silences in Japanese animated movies are sometimes replaced with words for
the purposes of explanation, clarification, and naturalization in the US versions
of the films.

From an acoustic point of view, it is often pointed out that some of the original tunes in
Miyazaki’s movies are recomposed in an effort to make the films more accessible to US
audiences. For example, Joe Hisaishi, the composer for Tenku no Shiro Rapyuta, arranged
his own original soundtrack with a symphony orchestra and extended it for the US
English version (Castle in the Sky), making it more than twice as long as the soundtrack
for the original Japanese version. Hisaishi (2002) increased the number of tracks from 14
Downloaded by [University of Tehran] at 05:50 17 February 2016

to 23 for the release of Castle in the Sky in the United States and extended the soundtrack
‘from 39 minutes to 90 minutes to reassure American audiences who expected animation
accompanied by music’ (Tomos, 2013, p. 210). The second English version of Majo no
Takkyubin (Kiki’s Delivery Service) clearly has more sound effects and incidental music
than the original, with the background music continuing to play for the first several
minutes without a break.
There is a strong presumption that, as a result of adding new sound effects and music
to Miyazaki’s films, silences in the original may be reduced in number and length.
However, apart from the obvious examples mentioned in the previous paragraph, there is
little agreement based on positive evidence as to how sound effects and background
music are dubbed in the US English version. Therefore, it is worthwhile focusing on
nonverbal acoustic elements among the four dominant multimodal components of AVT
(i.e. verbal visual, nonverbal visual, verbal acoustic, and nonverbal acoustic). The aim of
the present study is to examine silences not only as the absence of words but also as the
absence of sound.

4. Procedure
This study began by considering the first research question: How does the US English
translation of Spirited Away localize silences? In an attempt to answer this question, the
author compared the Japanese and US versions and examined the different ways in which
silence is treated in each.
The author then endeavored to answer the second research question: Is the US
English version of Spirited Away a global standard for how silences are dubbed? The
scope was expanded by comparing the US English translation to translations in five other
languages to determine whether the US version is typical in regard to the use of silences.
The following is a brief account of the procedure used in an attempt to answer the
first research question. The sound data were collected from the DVD of Spirited Away
released in the United States in 2003. The procedure for obtaining basic data involved the
following four steps.
First, audio files of the English and Japanese films were downloaded from the DVD
released in the United States. To prevent the distortion of data caused by differences in
format, only Dolby AC-3 files were adopted as basic material for further analysis. Next
the audio levels of those files were standardized using the ‘Loudness’ function of the
digital audio editing program Adobe Audition®, which set both files to a specific volume
 Perspectives: Studies in Translatology 147

level based on statistics. Adobe Audition® then marked the silences in each file
automatically, showing the number, starting time, and duration of each silence. Finally,
based on the results of the numerical analysis, specific examples of the English dubbing
were carefully examined and compared with the original Japanese version.

5. Results and discussion

5.1. Comparison of the Japanese and English versions
5.1.1. Silences lost in the process of translation
Table 1 shows the results of the search for silences in the Japanese and English versions
of Spirited Away. The places at which silences occur are indicated by the starting time and
the duration time. There are 31 silences in the Japanese language version and 21 silences
in the US English version. The numbers 1–10 in the column ‘Lost silences’ on the far
Downloaded by [University of Tehran] at 05:50 17 February 2016

right-hand side of Table 1 indicate the 10 silences that are lost in the English version. In
the process of translation into English, the number of silences dropped by 32.3%.

Table 1. Silences in the Japanese and US English versions of Spirited Away.

Japanese US English Lost silences

Starting time Duration Starting time Duration

0:00:000 0:10:187 0:00:000 0:10:258
3:32:851 0:06:131 3:32:945 0:05:699
5:16:475 0:05:017 1
14:34:936 0:05:641 2
34:07:613 0:11:725 34:07:652 0:11:635
34:25:124 0:14:051 34:25:163 0:13:959
34:48:868 0:06:293 34:48:820 0:06:289
44:09:070 0:29:977 44:09:063 0:18:240
44:40:979 0:14:373 3
44:58:959 0:09:397 44:58:458 0:08:984
45:08:779 0:07:170 4
47:37:493 0:07:181 47:35:410 0:05:258
1:04:46:107 0:05:892 5
1:05:04:124 0:06:374 1:05:04:528 0:05:950
1:05:18:610 0:09:698 1:05:19:349 0:09:139
1:07:06:249 0:05:046 6
1:07:46:516 0:05:599 1:07:46:736 0:05:671
1:08:29:166 0:05:385 7
1:10:30:843 0:05:989 8
1:11:51:001 0:18:220 1:11:52:824 0:05:759
1:12:16:742 0:05:212 1:12:15:146 0:06:750
1:20:36:924 0:06:499 1:20:36:973 0:06:313
1:28:19:421 0:07:393 1:18:19:291 0:05:035
1:42:47:394 0:05:737 1:42:47:451 0:05:549
1:42:58:843 0:13:302 1:42:58:788 0:12:438
1:45:17:036 0:05:103 9
1:45:55:591 0:07:827 1:45:55:744 0:07:408
1:46:01:853 0:08:034 1:46:01:538 0:08:005
1:55:11:992 0:09:673 1:55:11:773 0:09:835
1:56:31:464 0:06:669 10
2:04:21:892 0:14:683 2:04:22:356 0:14:219
 148 R. Adachi

The loss of 10 silences prompts the following question: How are these silences treated
during the process of translation? When translating Spirited Away, Cindy and Don Hewitt
(An Interview, 2003), the script adapters on Disney’s recent Miyazaki movie dubs, stated
that their objective was ‘to make the experience of watching the English dub as true as
possible to the experience which Miyazaki intended’. Cindy and Don Hewitt worked with
John Lasseter, executive producer of the English adaptation, who was responsible for the
whole dubbing project, including casting, and supervised the dubbing process that
proceeded in close consultation with Disney and Studio Ghibli. In an attempt to obtain a
workable compromise between conflicting goals, namely faithfulness and intelligibility,
three kinds of adaptations of silences were employed: replacement of lines, insertion of
fillers, and increase in sound effects.
Downloaded by [University of Tehran] at 05:50 17 February 2016

5.1.2. Replacement of lines

Some of these silences are filled with words for the purpose of explicitness,
domestication, and avoidance of awkwardness. There are 47 lines in the US English
version of Spirited Away that are not in the original Japanese script (Tamura, 2010, p.
93).5 A typical example is Lost silence 8, found in Chihiro’s monologue. Table 2 displays
the excerpts from the Japanese and US English versions.
It is evident that Cindy and Don Hewitt (An Interview, 2003) executed their
translation on a line-by-line basis. Most of the lines are translated faithfully, except for
the third line in the US English version, which is new: ‘I sure hope they’re doing all right
over there’. This additional line helps make Chihiro’s anxiety about her parents’ safety
more explicit (Yamada, 2004, p. 203). Moreover, the insertion of this new line allows the
dialogue to flow smoothly. The silence of the original Japanese version is skillfully
removed by elaborating on Chihiro’s previous line (‘There’s mom and dad’s pigpen’),
which is heard as a voice-over, defined here as the voice of an unseen character or of an
on-screen character who is not seen speaking, in the US English version.

Table 2. Excerpts from Chihiro’s monologue.

Japanese US English

CHIHIRO: CHIHIRO:
(1) A hontoni umi ni natteru Wow, it really is the sea.
Wow really sea has turned into There’s mom and dad’s pigpen.
Wow, it really has turned into an ocean
(Silence)
(2) Koko kara otosantachi n toko mieru n da I sure hope they’re doing all right over
Here from Hon-e.father-PLUR ATTR place see there.
ASS
I can see where my parents are from here.
(3) Kamaji ga mo hi wo taiteru Kamajii lit the boiler already?
Kamaji SUBJ already fire OBJ has lit
Kamajii has already lit the fires.
(4) Sonnani nechatta no kana How long was I asleep?
So long slept-end-up ASS QUEST (1:10:28–1:10:47)
Did I sleep so long?
Note: Hereinafter, interlinear gloss abbreviations are indicated as follows: AGENT, marks agent in passive
sentences; ASS, assertive marker; ATTR, attributive marker; HON, honorific; IMP, imperative form; OBJ, object
marker; PLUR, plural suffix; QUEST, question marker; SUBJ, subject marker; TOP, topic marker.
 Perspectives: Studies in Translatology 149

Table 3. Excerpts from the final scene.

Japanese US English

MOTHER: MOTHER:
(5) Chihiro hayaku shi nasai Come on, Chihiro. Let’s get to our new home.
Chihiro quickly do IMP FATHER: You’re not scared, are you?
Chihiro, hurry up. MOTHER: Don’t be afraid, honey. Everything’s going to be okay.
(End of dialogue) FATHER: A new home and a new school. It’s a bit scary.
CHIHIRO: I think I can handle it.
(2:00:20–2:00:50)

The English translation of Spirited Away has 16 lines in total that are added as voice-
overs. The frequent use of this adaptation technique seems to be characteristic of this US
Downloaded by [University of Tehran] at 05:50 17 February 2016

version. One of the most memorable examples of filling silence with voice-over dialogue
in the US English version occurs at the end of the movie. In the final scene, Chihiro and
her parents are riding in a car that is speeding away from the supernatural world and
transporting them to their new home. As Table 3 shows, there is no dialogue in the final
scene of the original Japanese version. By contrast, the ending of the US English version
features a conversation among the family members. Cindy and Don Hewitt (A Second
Interview, 2005), sensing that the original ending was so abrupt that ‘Chihiro’s emotional
journey did not seem complete’, decided to include several new lines at the end of the
movie. This additional dialogue at the end of the US English version strongly emphasizes
the family ties and Chihiro’s spiritual growth.
There is a somewhat different example in which the lines are changed in Lost silence 9.
As shown in Table 4, Yubaba’s brutality is blatantly obvious in the Japanese version but is
not discernible in the US version. All the original lines are replaced with a statement of the
conditions for freeing Chihiro and her parents from the supernatural world. Whereas the
conditions are ambiguous in the original version, the English translation clarifies the
ambiguity for audiences so that they cannot fail to understand what is going on (Yamada,
2004, p. 200). In addition, expressions of violence, such as ‘torn limb from limb’, that are
spoken in the Japanese version are not present in the US English version.6 This decision to
exclude such expressions in the US English version is presumably related to concerns raised
in the United States about the impact of media violence on young audiences. As a result, the
dialogue is extended for several seconds and fills the silence that follows in the original.

Table 4. Excerpts from Yubaba’s answer to Haku’s request.

Japanese US English

YUBABA: YUBABA:
(6) Sorede omae wa do naru n dai Fine! But on one condition!
And you TOP how become ASS QUEST
What will happen to you after that?
(7) Sono ato watashi ni yatsuzakinisarete mo ii n kai I get to give Sen one final test!
That after me AGENT be torn from limb to limb even If she fails, she’s mine. (1:45:10–
good ASS QUEST 1:45:25)
Is it okay even if you are torn from limb to limb
after that?
(Silence)
 150 R. Adachi

5.1.3. Increase in sound effects

It is characteristic of the US English version that insertions of exclamations occur
frequently during moments of silence. Some typical examples found in Spirited Away are
interjections, such as sobbing in Lost silence 3. It is early in the morning, and all the
employees of the bathhouse, except for Chihiro, are still asleep together in a large room.
In the Japanese version, Chihiro, with bated breath, trembles under her futon. Whereas
the Japanese version minimizes the sound effects – only employees’ breathing and birds’
chirping are faintly heard in the background – the US English version adds the distinct
sound of Chihiro’s helpless sobbing to the scene. This sound effect implies that she has
spent the night in tears. Clearly, this modification affects Chihiro’s characterization. In the
original, she is cowardly and clumsy, but she is not a blubberer. Separated from her
parents, she endures the difficult situation and seldom cries. In fact, she cries loudly only
once during the movie. In an early scene in the Japanese version of the film, Chihiro cries
Downloaded by [University of Tehran] at 05:50 17 February 2016

when she is moved by the unexpected kindness of Haku, a dragon who makes a rice ball
for her. Chihiro’s sobbing is added to the US English version in an attempt to make
Chihiro a more natural and easily relatable character to US audiences.

5.1.4. Insertion of pause fillers

Silences are considerably removed by making an existing sound louder or by adding new
sound effects and music. In Lost silence 2, for example, Haku is searching the backstreets
secretly for Chihiro. Chihiro is too scared to come out of the darkness from behind an old
house. In the original Japanese version, silence falls while the camera follows Haku
through the dark and empty backstreets. In the US English version, however, the silence
is broken by unidentified voices in the background and by the sound of Haku’s footsteps,
which have been obviously amplified.7 These sound effects function to avoid an awkward
silence in a dark and quiet scene that is monotonous aurally as well as visually, something
considered uncomfortable to US audiences. It is likely that audiences from low-context
cultures such as the United States tend to operate on the principle of valuing messages
that are direct and transparent in meaning. Although the silence may help make the scene
transition smoother, it does not highlight any dramatic effects. A dramatic silence, which
often occurs before or after something important is said or done, creates an intense
atmosphere, building up suspense or magnifying the importance in the moments of
meaningful tension or lingering resonance. However, the silence in this scene seems
almost devoid of semiotic content, which US audiences are unlikely to be accustomed to
and expect from animated movies.
Another example is Lost silence 5, the episode of the stink spirit, which reflects
Miyazaki’s environmentalist message. The spirit is muddy and covered with garbage and
pollution. After Chihiro washes the spirit clean in the bathhouse, it is, however, revealed
to be a splendid river spirit. When the river spirit carries Chihiro and gives her a special
dumpling as a token of his gratitude, silence reigns for nearly 10 seconds in the original
Japanese version. This is a lull that follows a string of scenes full of frenzied excitement
involving the washing of the stink spirit. The silence is symbolic of a sacred moment and
space in which the river spirit makes its appearance. However, the US English version
abolishes the silence by adding the sound effect of gentle popping which is synchronized
with the image of bubbles rising in a stream around Chihiro in the water. The relationship
between sound and image seems to be prioritized in the US English version. This sound
effect was probably added in the final editing to make the mysterious and unfamiliar
quality of this scene more natural to US audiences.
 Perspectives: Studies in Translatology 151
Downloaded by [University of Tehran] at 05:50 17 February 2016

Figure 1. Categories and numbers of omitted silences.

5.1.5. Classification of the lost silences

In Figure 1, the 10 instances of omitted silences in the US English version are grouped
into one of three categories: replacement of lines (linguistic), insertion of fillers
(paralinguistic), and increase in sound effects (acoustic). As a result, there are two
replacements of lines, three insertions of fillers, and five increases in sound effects.
Therefore, among the three ways in which silences are omitted, the increase in sound
effects is used the most frequently to avoid silences in Spirited Away.
As shown in Figure 1, the US English version of Spirited Away uses acoustic
adaptations to a greater extent than linguistic modifications to fill the silences in the
original Japanese version. This result is consistent with a study by Adachi (2012a),
showing that as Japanese animation grows in popularity, large-scale modifications of
dialogue in Miyazaki’s films become less common. Although fewer changes are made to
dialogue, silences are still avoided in Spirited Away through acoustic changes, for
example by inserting fillers and interjections.
Because nonverbal components, including exclamations and sound effects, are not
always written in scripts, they tend to be overlooked in analyses of linguistic data. Based
on the high frequency in the increase of sound effects in the US English version, as
previously mentioned here, it is important to pay more attention to the adaptation of
nonverbal components in the process of translation and to study AVT not only from the
linguistic perspective but also from the acoustic point of view.

5.2. Comparison of the number of silences among various translations

To investigate whether the US method of removing silences is typical (or not) among a
group of AVTs of Spirited Away in various parts of the world, a quantitative inquiry was
 152 R. Adachi

Table 5. Versions of Spirited Away.

Country Language Title Release date

Taiwan (ROC) Mandarin 神隠少女 (Shen Yin Shao Niu) [The Spirited- July 2002
Chinese Away Girl]
France French Le Voyage de Chihiro [The Voyage of Chihiro] November 2002
South Korea Korean 센과 치히로의 행방불명 (Sen-gwa Chihiroui December 2002
haengbangbulmyeong) [Sen and Chihiro Are
Missing]
USA English Spirited Away April 2003
Germany German Chihiros Reise ins Zauberland [Chihiro’s Travel November 2003
in Wonderland]
Czech Czech Cesta do Fantazie [Journey to Fantasy] December 2003
Downloaded by [University of Tehran] at 05:50 17 February 2016

conducted into the number of silences in each translated version. Dolby AC-3 files were
collected from the DVDs of Spirited Away released in the following six countries and
regions: Taiwan (Republic of China), France, South Korea, the United States, Germany,
and the Czech Republic (listed in order of release date). All these foreign versions are
dubbed as well as subtitled. Table 5 shows a list of the dubbed versions.
The volumes of sound data were adjusted in the same manner as measured in the US
English version. The number of silences (as previously defined) was counted automat-
ically for each translated version. The results are shown in Table 6.
There are marked differences in the number of silences among the translations under
investigation. Among these six foreign-language versions, the US version notably has the
least number of silences. The number of silences in the US version is nearly one-third of
those in the Chinese version and half of those in the French version. As mentioned earlier,
the US version adds many lines as voice-overs. This technique is not used in any of the
other AVTs, not even in the German version, which has the second-least number of silences
(Umezaki, 2007, p. 73). Viewed in this light, the US version of Spirited Away is unique and
cannot be regarded as a typical translation of the silences that are characteristic of Japanese
animation.
This point deserves explicit emphasis. English is often used as a pivot language, or a
bridge language – that is, an interlingua – for translation from Japanese to many different
languages (Díaz Cintas, 2009, p. 10).8 The advantage of using English as a pivot language
is that pivot translations using English are more cost-effective than direct translations from
source languages to target languages. American editions of Japanese animated films have a
strong influence on the AVT of Japanese animation into other languages in the global
market. The findings of this study, however, suggest that the US English version of Spirited

Table 6. Number of silences in different translations.

Country (language) Number of silences

Taiwan (Mandarin Chinese) 60

France (French) 44
South Korea (Korean) 35
Czech Republic (Czech) 35
Germany (German) 26
United States (English) 21
 Perspectives: Studies in Translatology 153

Away is by no means culturally neutral. So, if this version is used as a pivot for other
translations, there exists a risk of loss of quality in the process of double translation,
whereby a source text is translated first into an intermediate language (usually English) and
then into other, less widely used languages.9 There is a risk that the double translation can
compound the problems already produced in a single translation, such as ambiguities,
inaccuracies, and errors, especially when the individual translating the material from the
intermediate language into the target language does not understand the source language
(Edström, 1991).

6. Conclusions
The following two conclusions are drawn from this study’s results.

(1) The US English version of Spirited Away removes silences by inserting fillers
Downloaded by [University of Tehran] at 05:50 17 February 2016

and by adding or amplifying sound effects. Acoustic adaptations are greater in

number than lexical ones. This finding suggests that it is important to study AVT
not only from the verbal perspective but also from the acoustic point of view.
(2) Among the six foreign-language versions, the US version is unique in its marked
tendency to avoid silences. Thus, the US version should not be considered a
global standard, even though US AVTs often exert substantial influence as an
interlingua in the dubbing process into other languages.

Because sound has such a powerful influence and can make a direct appeal to the emotion
of audiences, it is important not to overlook the added sound effects and the high
frequency of sound modifications in the US English version of Spirited Away. Generally
speaking, restrictions in the process of AVT are less rigid on acoustic elements than on
lexical ones. It is easier and less obstructive to manipulate sound effects and background
music than to add or reduce words in synchrony with the number of syllables and lip
movements of the characters on-screen. This observation is consistent with the fact that
the removal of silences in Spirited Away was executed mainly with acoustic transactions,
including the insertion of pause fillers and sound effects. Therefore, a broader
understanding of the acoustic aspects in the process of translation and dubbing will
help develop a more comprehensive view of AVT.

Acknowledgements
This paper owes much to the thoughtful and constructive comments of the anonymous referees.

Funding
This work was supported by the Japan Society for the Promotion of Science (JSPS) under
KAKENHI [grant number 23520355].

Notes
1. Official English titles of Japanese animated works are shown in italics in the text as they are or
in parentheses, and word-for-word translations of original Japanese into English are provided
within square brackets.
2. It is crucially important in the process of automatic dialogue replacement (ADR) to ensure that
translated dialogue is post-dubbed properly in synchronization with the picture. Although
synchronizers or adapters are directly responsible for matching translated dialogue to the mouth
movements and facial expressions of animation characters, the task may be carried out by
 154 R. Adachi

dubbing directors who are in charge of all modifications of the translated script written by
translators (Martinez, 2004, p. 6).
3. The term ‘dialnorm’ is an abbreviation for the term ‘dialogue normalization’. It is a parameter
within the Dolby Digital (AC-3) system which identifies the area of normal speech in an audio
program.
4. Adachi (2010) attempted to keep the sample of movies as homogeneous as possible by adopting
five criteria for selecting translations: country, the year of release of a translated version, genre
keywords on the Internet Movie Database (IMDb), motion picture rating or certification by the
Motion Picture Association of America (MPAA), and box-office records obtained from Box
Office Mojo.
5. The US English translation of Hauru no Ugoku Shiro (Howl’s Moving Castle) has 26 lines that
are not in the original script, and Mononoke Hime (Princess Mononoke) has 65 lines that are not
in the original script.
6. Spirited Away is released as G (general, suitable for all ages) in Japan but is rated as PG-13
(parental guidance suggested if under 13 years of age) in the United States for some scary
moments.
Downloaded by [University of Tehran] at 05:50 17 February 2016

7. Yamada (2004) observed that, as far as the sound effects are concerned, the dubbed English
version of Spirited Away is almost entirely faithful to the original, with the only possible
exception being the amplified voices and footsteps in this backstreet scene (p. 197). However, as
the results of this present study show, there are many other modifications of sound effects made
in the US English version. K. Yamada’s inaccurate view indicates that it is difficult to perceive
correctly how acoustic manipulations occur in the process of AVT.
8. With the exception of some popular series produced by major Japanese developers and
publishers, Japanese video games are usually translated from the US English versions into
European languages rather than from the original Japanese versions (O’Hagan, 2009, 2012).
9. A good example of this is the first Czech version of Spirited Away. Following the US English
version released in the United States in April 2003, the first Czech version was produced by
Intersonic in December 2003. In November 2004, however, the second Czech version of Spirited
Away was distributed by hypeMAGAZINE. In this revised version, Czech subtitles are corrected
based on a direct translation from the Japanese version.

Notes on contributor
Reito Adachi is a professor of English Language and Literature at Kurashiki City College, Japan.
He holds MAs from Okayama University, Japan, and Hollins University, USA, and received his
PhD degree from Okayama University. His current research interests include audiovisual translation
and children’s literature.

ORCID
Reito Adachi http://orcid.org/0000-0001-8563-3576

References
Adachi, R. (2010). Basic analysis of dialogue data on pauses and silences in Japanimation movies:
Comparison of original Japanese and American animations with their translated versions.
Eibeibunka, 40, 243–259.
Adachi, R. (2012a). A diachronic comparison of the English translations of Hayao Miyazaki’s
animated works. Eibeibunka, 42, 97–118.
Adachi, R. (2012b). A study of Japanese animation as translation: A descriptive analysis of Hayao
Miyazaki and other anime dubbed into English. Boca Raton, FL: Dissertation.com.
Adobe Audition® (CS6) [Computer software]. Retrieved from http://tv.adobe.com/product/audition/
Bogucki, Ł. (2009). Amateur subtitling on the Internet. In J. Díaz Cintas & G. Anderman (Eds.),
Audiovisual translation: Language transfer on screen (pp. 49–57). Basingstoke: Palgrave Macmillan.
Bosseaux, C. (2008). Buffy the vampire slayer: Characterization in the musical episode of the TV
series. The Translator, 14, 343–372. doi:10.1080/13556509.2008.10799262
Chaume, F. (2012). Audiovisual translation: Dubbing. Manchester: St Jerome.
 Perspectives: Studies in Translatology 155

Delabastita, D. (1990). Translation and the mass media. In S. Bassnett & A. Lefevere (Eds.),
Translation, history and culture (pp. 97–109). London: Cassell.
Díaz Cintas, J. (2009). Introduction – audiovisual translation: An overview of its potential. In J. Díaz
Cintas (Ed.), New trends in audiovisual translation (pp. 1–18). Bristol: Multilingual Matters.
Díaz Cintas, J., & Muñoz Sánchez, P. (2006). Fansubs: Audiovisual translation in an amateur
environment. Journal of Specialised Translation, 6, 37–52. Retrieved from http://www.jostrans.
org/issue06/art_diaz_munoz.pdf
Ebert, R. (2002). Hayao Miyazaki interview. Retrieved from http://www.rogerebert.com/interviews/
hayao-miyazaki-interview
Edström, B. (1991). The transmitter language problem in translations from Japanese into Swedish.
Babel, 37(1), 1–14.
Elwood, K. (2003). A comparative analysis of requests in Majo no Takkyubin and Kiki’s Delivery
Service. Bunkaronshu, 22, 213–236.
Gonzalez, I. (2009). Culturemes in Mononoke Hime: A comparative analysis of subtitles (English–
Spanish; DVD–Internet). CineMagaziNet!, 13. Retrieved from http://www.cmn.hs.h.kyoto-u.ac.
jp/CMN13/irene-article-2009.html
Downloaded by [University of Tehran] at 05:50 17 February 2016

Hisaishi, J. (2002). Castle in the sky: US version soundtrack [CD]. Tokyo: Tokuma Shoten.
An Interview with Cindy and Don Hewitt [Interview transcript]. (2003, December). Retrieved from
Hayao Miyazaki Web site http://www.nausicaa.net/miyazaki/interviews/hewitt_interview.html
Kusanagi, S. (2003). Amerika de Nippon no anime wa do mirarete kitaka [How have Japanese
animations been seen in America?]. Tokyo: Tokuma Shoten.
Leonard, S. (2004). Progress against the law: Fan distribution, copyright, and the explosive growth
of Japanese animation. Cambridge, MA: Massachusetts Institute of Technology. Retrieved from
http://web.mit.edu/seantek/www/papers/progress-columns.pdf
Martinez, X. (2004). Film dubbing: Its process and translation. In P. Orero (Ed.), Topics in
audiovisual translation (pp. 3–7). Amsterdam: John Benjamins.
O’Hagan, M. (2009). Putting pleasure first: Localizing Japanese video games. TTR: Traduction,
Terminologie, Rédaction, 22(1), 147–165.
O’Hagan, M. (2012). Transcreating Japanese video games: Exploring a future direction for
translation studies in Japan. In N. Sato-Rossberg & J. Wakabayashi (Eds.), Translation and
translation studies in the Japanese context (pp. 183–201). London: Continuum.
Oittinen, R. (1993). I am me—I am other: On the dialogics of translating for children (Doctoral
dissertation). University of Tampere, Tampere.
Okuhara, R. (2009). The censorship of Japanese animé in America: Do American children need to
be protected from Dragon Ball? In M. I. West (Ed.), The Japanification of children’s popular
culture: From Godzilla to Miyazaki (pp. 199–208). Lanham, MD: Scarecrow.
Palencia Villa, R. M. (2002). La influencia del doblaje audiovisual en la percepción de los personajes
[The influence of the audiovisual dubbing on the perception of the characters] (Doctoral
dissertation). Universidad Autónoma de Barcelona, Barcelona.
Pérez-González, L. (2006). Fansubbing anime: Insights into the ‘butterfly effect’ of globalization on
audiovisual translation. Perspectives: Studies in Translatology, 14(14), 260–277. doi:10.1080/
09076760708669043
Poyatos, F. (1997a). Aspects, problems and challenges of nonverbal communication in literary
translation. In F. Poyotas (Ed.), Nonverbal communication and translation: New perspectives
and challenges in literature, interpretation and the media (pp. 17–47). Amsterdam: John
Benjamins.
Poyatos, F. (1997b). The reality of multichannel verbal-nonverbal communication in simultaneous
and consecutive interpretation. In F. Poyotas (Ed.), Nonverbal communication and translation:
New perspectives and challenges in literature, interpretation and the media (pp. 249–282).
Amsterdam: John Benjamins.
Poyatos, F. (2002). Nonverbal communication across disciplines, volume II: Paralanguage,
kinesics, silence, personal and environmental interaction. Amsterdam: John Benjamins.
A Second Interview with Cindy and Don Hewitt [Interview transcript]. (2005, May). Retrieved from
Hayao Miyazaki Web site http://www.nausicaa.net/miyazaki/interviews/hewitt_interview2.html
Tamura, C. (2010). Nippon no animeshon eiga no eigo fukikaeban ni okeru serifu no tsukekuwae ni
kansuru kenkyu [A study of the addition of lines in the dubbed versions of Japanese animated
films]. Invitation to Translation Studies, 4, 87–105.
 156 R. Adachi

Teshigawara, M. (2003). Voices in Japanese animation: A phonetic study of vocal stereotypes of

heroes and villains in Japanese culture (Doctoral dissertation). University of Victoria, Victoria,
British Columbia.
Tomos, Y. (2013). The significance of anime as a novel animation form, referencing selected works
by Hayao Miyazaki, Satoshi Kon and Mamoru Oshii (Doctoral dissertation). Aberystwyth
University, Aberystwyth.
Umezaki, M. (2007). Doitsu ni okeru nippon animeshon no eizohonyaku ni kansuru chosakenkyu
[Surveying and research of the audiovisual translation of Japanese animation in Germany]. In
Tokuma Memorial Cultural Foundation for Animation (Ed.), Tokuma Kinen Animeshon Bunka
Zaidan nenpo bessatsu [Separate-volume annual report 2006–2007: Tokuma memorial cultural
foundation for animation] (pp. 52–165). Tokyo: Tokuma Memorial Cultural Foundation for
Animation.
Williams, E. A., Jones, G. A., Layer, D. H., & Osenkowsky, T. G. (Eds.). (2007). National
association of broadcasters engineering handbook (10th ed.). Burlington, MA: Focal Press.
Yamada, H. (1997). Different games, different rules: Why Americans and Japanese misunderstand
each other. New York, NY: Oxford University Press.
Downloaded by [University of Tehran] at 05:50 17 February 2016

Yamada, K. (2004). Eigo-ban anime sakuhin ni miru honyaku no mondai: Sen to Chihiro no
Kamikakushi no baai [On problems of translation in English version of anime: A case study of
spirited away]. Journal of the Faculty of Global Communication, Siebold University of Nagasaki,
5, 195–205.