An Analysis of Translation Shifts of

English Email Letters in to Indonesian

by Google Translation

Edi Suprayetno
edisuprayetno@gmail.com

ABSTRACT
The objectives of this study were to (1) explain how the types of translation shifts are processed in English
email letters in to Indonesian by google translation, (2) investigate types of translation shifts, (3) investigate types
of grammatical shifts found in the translation of English email letters in to Indonesian by google translation at PT.
Mutiara Medica Sunggal during December 2014. Based on the analysis by Catford theory as well as the
characteristics of each type of translation shift it was found that translation shifts occured are due to the shifting of
grammatical unit as well as the small linguistic changes system occur between ST and TT in order to make the
target language is more natural and understandable by the readers. The types of translation shifts occured were (1)
stucture shift (97 shifts or 42,54%) of the total 228 shifts, (2) unit shift (78 shifts or 34,21%), (3) level shift (24
shifts or 10,53%.), (4) Intra system shift (23 shifts or 10,09%) , and(4) class shift (6 shifts or 2,63%). The
dominant tpye is structure shift. It means that in conducting the process of translation, google translation is more
effective when it translate in the level of word to word or phrase to phrase. The types of grammatical shifts occured
were (1) Phrase (Modifier + Noun) to Phrase (Noun + Modifier) (76 shifts or 33,33%), (2) shift of word to phrase
(51 shifts or 22,37%), (3). Shift of phrase to word (27 shifts or 11,84 %) , (4) shift of grammatical unit to lexicon
(24 shifts or 10,53 %) ,(5) shift of plural noun/ noun phrase to singular noun/ noun phrase (23 shifts or 10,09 %) ,
(6) pharse to phrase with different linguistic system from SL to TL (21 shifts or 9,21%), and (7) shifts of clause to
phrase, clause to word and clause to sentence which reach 3 shifts or 1,32%, 2 shifts or 0,88% and 1 shift or 0,44%
of the overall data. In conclution, that google translation is only effective when it is used to conduct the translation
from word to word, Phrase to Phrase, word to phrase and phrase to word but not in phrase to clause , clause to
clause or sentence to sentence.

Key words : Translation Shifts, English Email Letters, Google Translation

1.Introduction
This study is to analyze the types of translation shifts found in the received email English letters
in to Indonesian by google translation during December 2014 at PT.Mutiara Medica Sunggal as well as
the types of grammatical hirecachy occured on the shifts on translation and how are the types of
translation shifts processed in email English letters in to Indonesian by google translation. The company
is located at Jl.Pinang Mas III Blok G. No 4 Komplek Villa Palem Kencana Sunggal, where the writer
works as a freeline translator.
In the preliminary, search inside the received email English letters which is translated by google
translation in to Indonesian it is found some translation shifts in level of stucture, unit,class and intra
system. This preliminary findings is as a starting point for the reseacher to analyze the kinds of shift in
the translated email, for instance, it is found :
(1) SL: I am sorry for the late response
(1a) TL : Saya minta maaf untuk respon terlambat
From this translation process it is found that there is a shift from grammar to lexis in which the
patterns to be am (grammar) in the source language text is translated into lexicon minta in the target
language text. This shift is called level shift.
In connection with the category shifts, it is also found that :
(2) SL : Long time not receive your information.
(2a) TL : lama tidak menerima informasi anda.
From the example, it found that the noun phrase in the SL change the form in to adjective in the
TL. This shift is called as class shift.
From the preliminary findings the reseacher is very much intersted in conducting further analysis
to google translation product of received email by PT. Mutiara Medica, Sunggal especially the one that is
choosen for this thesis which focuses on the types of translaion shifts.
Based on the shifting of structure concept, there is a basic technique to solve the problem in
shifting of the translation, that is Translation shift, the replacement of one grammatical unit by another.
For example: catalogues translated into katalok. In this case the transposition that a change happens from
plural noun into singular noun. Transposition is the only translation procedure concerned with
grammatical. In translating text from source language into target language, transposition is main process
in the translation that should be taken into account by translator.
Nida (1964) in Venuti (2000 : 126) says that there can be no absolute correspondence between
languanges since no two languanges are identical. Hence, it can be concluded that due to no two identical
languanges, shifts may occur in translation. Catford (1965) in Venuti (2000 : 141) states that shifts mean
the departures from formal correspondence in the process of going from SL to TL. Catford says that there
are two major types of shifts occur. They are level shifts and category shifts. Level shifts is a shifts from
grammar to lexis. It means that a grammatical unit in English, such as noun, affix, etc, has a lexical unit in
Bahasa Indonesia as its translation equivalent (Machali, 1998 : 14).
Category shift is about unbounded and rank-unbounded translation. In unbounded translation
equivalences are not tied to a particular rank and may additionally find equivalences at sentence clause
and other levels, while in rank-unbounded translation an equivalent is sought in the target language (TL)
for each word or for each morpheme encountered in the source language (SL) . This thesis focuses on
category shifts. These category shifts have four kinds, namely structure shifts, class shifts, unit shifts, and
intra-system shifts.

1.1 The Problems of the Study

The focus of the study is to find out the level shift and the category shift of translations. These
two kinds then alaborated in to more specific reseach questions as follows :
1) How are the types of translation shifts processed in English email letters in to Indonesian by google
translation ?
2) What types of translation shifts occured in the translation of English email letters in to Indonesian by
google translation ?
3) What types of grammatical shifts found in the translation of English email letters in to Indonesian by
google translation ?

2. Review of Literature

Translation theory was influenced by Nida (1964), a bible translator from United States, and
Catford (1965) who strongly advocated the central concept of equivalence in contrast to traditional
dichotomy of faithful versus free. Bell (1991) supported their view by stated the definition of translation
as the expression in another language or (target language) of what has been expressed in another, source
language.
Bell (1992: 5-7) describes that translation should preserve the equivalence of the text from the
Source Language Text (SLT) into the Target Language Text (TLT). Baker (1992) explains that in
translation there are mainly two kinds of equivalence, structural equivalence and textual equivalence. But,
by different experts the components of this equivalence can be found different. Dubois (1973) in Bell
(1991) describes that a translation should be preserving semantic and stylistic equivalence. Other
translation scholars, Hartman and Stork (1972) in Bell (1992) group three kinds of approach in
equivalence, that is, (1) equivalent in different degrees, (2) equivalent in respect of different levels of
presentation and (3) equivalent at different ranks.
Based on all definitons above, it can be concluded that translation is a communication process
which involves the transfer of a message from SL to TL. Translation then consist of studying the
lexicon,grammatical structure and communication situation as well as the cultural context of the source
language text, analysing it in order to determine its meaning, and then reconstructing this same meaning
using the lexicon and grammar structure which are appropriate in Target Language (TL) and its cultural
context.

The Process of Translation

Tranalstion is not just transfer word by word from SL to another language. This can not guarentee
quality of translation . Translation is not only concentrated on the word only. In other words translation is
not only simply reflacing words with similar meaning but of finding appropriate ways of saying in other
language.For accurate translation from SL to the TL contextual diction in other language is one of the
most important te be considered
Newmark (1998 : 19 30) describes the procedures of translation, where it is begins with the
choosing of method of approach and secondly is level of translation : (1) the SL text level; (2) referential
level; (3) the cohesive level; (4) the level of naturalness.
According to Newmark(1998 : 21) states that there are two approaches to translating and many
compromises between them : (1) sentence by sentence; (2) read the whole text then find the intention,
register, tone, mark the difficult words and passages, then start translating.

Types of Translation
Jacobson (1959) in Venuti (2000:114) clasified type of translation into three as there are : (1)
intralingual translation; (2) interlingual translation and; (3) intersemiotic trasnlation.
Intra lingual translation is an interpretation of verbal sign by means of the other sign of the same
language. In this type, the translator uses different words and phrases to communicate similar meaning.
This type will occur, when the translator want to paraphrase an expression or text in the same language as
in 1) SL : Bubur, TL : Nasi yang dimasak sampai benar benar lunak; 2) SL : Breakfast, TL : first meal
of the day.
Interlingual translation is an interpretation of verbal signs by means of some other language. This
converts a text written in the source language (SL) to one written in the target language (TL). In
conclution in this type, the translator changes the content of an original written text from one language
properly in to the written text of another language, as english to Bahasa Indonesia, as in SL : I went to
Jakarta yesterday, then translated in to TL : Saya pergi ke Jakarta Kemaren.
Intersemiotic translation is an interpretation of verbal signs by means of nonverbal sign system.
Intersemiotic includes the interpretation of a text in to another system sign. In other words, the translation
involves the changing of a sign in to another system of sign. It occurs when a person want to translate an
artifact or symbol.
Example :

SL :

TL : No stop here
Based the explaination above, it can be concluded that each types of translation has its own
special fuction and characteristics. Intralingual fouses on the translating a text in the same language. Then
interlingual translation focuses on the translating a text from one language in to another language, and
intersemiotic translation focuses on the translating a sign in to another sign. These types will be very
important to be considered by a translator when conducting a translation.

Shifts in Translation
Catford (1965) defines shifts as departures from formal correspondence in the process of going
from the Source Language to the Target Language.Then Hatim and Munday (2004:26) state that
translation shifts is the small linguistic changes that occur between ST dan TL. John Catford was the first
scholar to use the term in his A Lingustic Theory of Translation (1965). There are two kinds of translation
shifts:

Level shift
 Level shift means that a source language item at one linguistic level has a target language
translation equivalent at a different level. As pointed by Catford, it may occur because the translation
between these levels of phonology and graphology or between these levels and the levels of grammar and
lexis is impossible. Translation between these levels is absolutely ruled out by the theory, which posits
relationship to the same substance as the necessary condition of translation equivalence. Then, with shifts
from grammar to lexis or vice versa as the only possible level shifts in translation; and such shifts are, of
course, quite common. For example: ST : We are producing now . SL : Kami sedang memproduksi
sekarang
In this translation, there is a shift from grammar to lexis in which the patterns to be +v-ing
(grammar) in the source language text is translated into lexicon sedang in the target language text.

Category Shifts
Catford refers it to the unbounded and rank-bound translation. The first being approximately
normal or free translation in which SL-TL equivalences are set up at whatever rank is appropriate.
Usually, but not always, there is sentence sentence equivalent, but in the course of a text, equivalences
may shift up and down the rank scale, often being established at rank lower than the sentence. Rank
bound translation is used only to refer to those special cases where equivalence is deliberately limited to
ranks below the sentence, thus leading to bad translation that is translation in which the TL text is either
not a normal TL form at all, or is not relatable to the same situational substance as them SL text.
However, it is clearly meaningless to talk about category shift unless we assume some degree of formal
correspondence between SL and TL.

Structure Shift
This category is the most frequent category shifts at all ranks in translation: they occur in
phonological and graphological translation as well as in total translation. For example : SL : New order is
translated as TL : Pesanan baru.
New order in the source language text is constructed of modifier (new) + Noun (order),
meanwhile in the target language it becomes Pesanan baru which is constructed of Noun (pesanan) +
modifier (baru).

Class Shift
For this term, Catford is following Hallidays definition . Class is defined as that grouping of in
the structure of the unit next above. Class shift occurs when the translation equivalent of a SL item is a
member of a different class from the original item. Because of the logical dependence of class on
structure, it is clear that structure shifts usually entails class shifts, though this may be demonstratable
only at a secondary degree of delicacy. For example:ST : Medical Equipment as in TL : Alat medis. In
this example, medical in the source language text is an adjective, meanwhile medis in the target language
text is a noun.

Unit Shift
This kind of shift involves change in rank. It departs from formal correspondence in which the
translation equivalent of a unit at one rank in the SL is the unit at a different rank in the TL. For example:
a phrase into a clause as in TL : Health care is translated in to ST : Memelihara kesehatan badan.

Intra System Shift

The terms intra-system shift is used for those case where the shift occurs internally, within a
system; that is, for those cases where SL and TL posses systems which approximately correspond
formally as to their constitution, but when translation involves selection of a non-corresponding term in
the TL system.Example : SL : Items is translated in to TL : Item. Therapies in SL in plural noun and it is
translated Terapi (singular) in TL.
Grammatical Hierarchy
In the point of translation shift above, the writer has explained about the level shift and category
shifts. In this section, the writer is going to recover further about grammatical hierarchy and rank scale.
As has been explicated before, unit means a stretch of language which is the carrier of a pattern of a
particular kind. The units of grammar operates in hierarchies, which is called as grammatical hierarchy,
larger units being made up of smaller or less inclusive units. They form a scale of units at different ranks,
also called as rank scale (Sulaiha :2010 :15-16)
There are five units in English grammar as they are : sentence, clause, group, word and
morpheme. The largest or highest on the rank scale is the sentence. The smallest or the lowest on the rank
scale is the morpheme.
As it is in Bahasa also has the similar grammatical hierarchy and units as well since the analysis
on translation shifts can only be done if there is a formal correspondence between the SL and the
TL.Bahasa also has five units of grammar, morfem (morpheme), kata (word), frasa (phrase/group), klausa
(clause), and kalimat (sentence). The study of grammatical hierarchy started to be applied in Indonesian
linguistics study about years 1960-1970s,(Rochayah Machali :1998 in Sulaiha : 2010 : 16)

Cause of Translation Shifts

Lexical Aspects
Meaning components packaged in the lexical items of a language are different to another
language. For example, the translation of the word projector in Bolivia is the thing that shows picture on
the wall.
The same meaning components can occur in several surface structure lexical items (forms). For
example, there is the word sheep in English, but the words lamb, ram and ewe also include the meaning
of the word sheep. They include the additional meaning components of young (in lamb), adult and male
(in ram), and adult and female (in ewe) (Larson : 1984)
A single meaning may be expressed in a variety of forms (Larson, 1984 : 6 - 9) For example, the
meaning the cat is black may be expressed by the following: the cat is black, the black cat, and the cat,
which is black depending on how that meaning relates to other meanings.

Grammatical Aspects
Parts of speech are language specific. Each language has its own division of the lexicon into
classes such as nouns, verbs, adjectives, etc.
Grammatical constructions also vary between the source language and the receptor language. The
order, for example, may be completely reserved (Larson : 1984 : 18 19). Larsons theory about this
cause of translation shift goes along with Catfords theory about translation shift itself

Google Translate
Google Translate soon proved to be very powerful, but with a great variation in the degree of
accuracy between individual languages. One recent study (Aiken and Balan 2011) showed that
translations between European languages were usually good, while those involving Asian languages
often relatively lacked in quality a fact closely tied to the (un)availability of large and qualified corpora.
GTs accuracy also seems to be text-type, genre or subject domain related. The Aiken 2009 study
established GTs usefulness in supporting large groups using up to 41 languages in a multilingual meeting
while some earlier studies (e.g. Hutchins and Somers 1992) showed the effectiveness of MT for weather
forecasts, or when working with simpler or standardised texts.

Conceptual Framework

Google Translation
Prosess Shifts of Translation
1. Level Shift
2. Category Shifts
Structure Shift
Class Shift
Unit Shift
Intra system Shift
 Grammatical Hierachy Data Source
1. Morpeme
2. Word Received English email
3. Phrase Letters
4. Clause
5. Sentence

Source Language (SL) Target Language (TL)

English Indonesian

English email letters translated in to

Indonesian by google translation

Types of Translation Shifts Analyzed by Grammatical Shifts

processed by Google Catford (1965) Theory processed by Google
Translation Translation

3. Research Design
The design of method used in this research is descriptive qualitative research to analyze
the translation shifts in translation product of received Engliah email letters at PT. Mutiara
Medica Sunggal during December 2014 as suorce text (ST), translated in to Bahasa by using
google translation as a target text (TT). Method and technique of analyzing data as well as
method and technique of presenting the analyses.

Source of Data
The data of this research were taken randomly from received English email letters
translated in to Indonesian by google translation at PT.Mutiara Medica Sunggal from 6 different
companies during December 2014. The English email letters are translated by using google
translation by the company itself in order to understand and comprehend the meaning of the
email in Indonesian.The data used in this study consist of 27 received English email letters
which were translated by google translation from 6 different medical companies from china.
The recieved English email letters were translated by google translation by PT. Mutiara Medica
in order to understand the meaning of the recieved emails then replied them by the help of
human translator. From the 27 received English email letters taken as data sources, all of them
are kinds of business letters as they are : offering letter, informing letter, asking information
letter, items replacement letter, etc.

Data Analysis
Lecompte and Schensul (1999 in Kawulich : 2004 : 97) define analysis as the process a
researcher uses to reduce data to a story and its interpretation. Data analysis is the process of
reducing large amount of collected data to make sense of them. In addition Patton (1987 in in
Kawulich : 2004 : 97) indicates that three things occure during analysis : data are organized, data
are reduced through summarization and categorization, and patterns and themes in the data are
identified and linked.
In order to gather the data needed in conducting the analysis of translation shift on
google translation, the writer refers Miles and Hubermans model (1984: 23 ) to analyze the data
descriptively. This model focused on four activities as they are : (1) collecting the data, (2)
reducing the data, (3) displaying the data, and (4) drawing a conclution.

4.Data Analysis, Research Findings and Discussion

Data Analysis
a. Level Shift

Data 6b
[35] SL : I am waiting for your early reply
[35a] TL : Saya sedang menunggu balasan awal Anda
In this process of translation, there is a shift from grammar to lexis in which the patterns
to be + v-ing [35] (grammar) in the source language is translated into lexicon sedang [35a] in
the target language .

Data 11c
[77] SL : I have to pay by myself.
[77a] TL : Saya harus membayar sendiri.
In this process of translation, the modal verb have to [77] in source languge is translated
in to lexicon harus [77a] in the target language .
Based on the catford theory, it can be concluded that level shift occurs in the process of
translation when the unit of grammatical rule or pattern in source language is translated in to
lexicon in target languge. In other words it is processed when one linguistic level has a target
language translation equivalent at a different level. As pointed by Catford, it may occur because
the levels of grammar and lexis.

b.Structure Shift

Data 14c
[99] SL :Thanks for your email.
[99a] TL :Terima kasih atas email Anda
The structure shift occured in data 14c [99] that your email in source language is
constructed of modifier (your) + Noun (email), meanwhile in the target language it becomes
email anda which is constructed of Noun (email) + modifier (anda).

Data 16d
[124] SL : There are no news from you after my last email
[124a] TL : Tidak ada berita dari Anda setelah email terakhir saya,
From the analysis of data [124] in 16d it is found that noun phrase of my last email in
the source language text is constructed of modifier (my - last) + Noun (email), is translated in to
email terakhir saya in the target language. The form is constructed of Noun (email) + modifier
(my - last). In other words the structure shift occurs when the noun phrase (modifier + noun) is
also translated in to noun phrase in the target language which is constituted out of noun phrase
(noun + modifier).

c.Class Shift
Data 4a
[20] SL : Please check the attached which difference
[20a] TL : Silakan periksa perbedaan terpasang
From the analysis of data [20] in 4a it is found that the adjective clause of of the
atteched which difference in the source language text , is translated in to adjective phrase
perbedaan terpasang. Therefore, it can be said that the is an occurance of translation shift in
this data in which adjective clause in source language is translated in to adjective phrase in target
language. There is no loss or gain of information in this translation.

Data 19d
[164] SL : please let me know your opinion about our products?
[164a] TL : silahkan beritahu saya pendapat Anda tentang produk kami?
Data [164] in 19d the verb clause let me know in source language is translated in to
verb phrase beritahu saya in target language by goole translation.

Data 20d
[178] SL : If it works, it means the patient plate cable
[178a] TL : Jika berhasil, berarti kabel pelat pasien
Data [178] in 20d the verb clause if it works in source language is translated in to
jika berhasil in target language by goole translation. It means that translation equivalent of a
source language item is a member of a different class from the original item in target language.

Data 22 e
[192] SL : would you please resend again the picture...
[192a] TL : tolong kirim ulang lagi gambar...
From the analysis of data [192] in 22e it is found that the verb clause of of would you
please in the source language text , is translated in to verb tolong. Therefore, it can be said
that there is an occurance of translation shift in this data in which clause in source language is
translated in to word in target language. There is no loss or gain of information in this translation.

4.Unit Shift
Data 9b
[53] SL : Received your email with many thanks
[53a] TL : Menerima email Anda dengan banyak terima kasih
Data [53] in 9b it was also found the noun thanks in source language is translated in
to noun phrase terima kasih in target language by goole translation. It means that translation
equivalent of a source language item is a member of a different unit from the original item in
target language.

5.Intra - System Shift

Data 5a
[25] SL : the last email with documents,....
[25a] TL : email terakhir dengan dokumen,....
From the analysis of data [25] in 5a it is found that the plural noun of documents in
the source language text , is translated in to singular noun dokumen. Therefore, it can be said
that there is an occurance of translation shift in this data in which plural noun in source language
is translated in to singular in target language. There is no loss or gain of information in this
translation.
The percentage of translation shifts of all data from received English email in to
Indonesian by google translation at PT. Mutiara Medica during December 2014 is shown in table
4.1 below :

Table 4.1 The Percentage of Types of Translation Shift Frequency in Received English
Email Letters in to Indonesian by Google Translation at PT. Mutiara Medica Sunggal
During December 2014
Types of Data
Translation A B C D E F f f%
Shift
Level Shift 0 4 8 7 3 2 24 10,53%
Sructure Shift 11 20 21 22 14 8 97 42,54%
Class Shift 2 0 0 1 3 0 6 2,63%
Unit Shift 7 14 13 13 20 11 78 34,21%
Intra System 8 1 0 7 1 6 23 10,09%
Shift
Total 228 100%

From the table avobe it is shown that from 228 data of translation shift in the 27 English email
letters in to Indonesian by google translation taken from 6 different companies during December 2014, it
was found that the dominant type of translation shift is stucture shift which reaches the highest total
frequency is 97 shifts or 42,54% of the total 228 shifts. The second level is unit shift with 78 shifts or
34,21% of the total data then followed by level shift which reaches 24 shifts of the overall data or
10,53%. The next level is Intra system shift which reaches 23 shifts of the overall data or 10,09%, and
the last level is class shift which only reaches 6 shifts of the overall data or 2,63%.

6.Grammatical Shift
As Catford (1965) in Venuti (2000 : 141) stated that shifts mean the departures from formal
correspondence in the process of going from SL to TL. Shift is the replacement of one grammatical unit
by another as weel as the shifting of structure concept. In another hands Hatim and Munday (2004 : 26)
also states translation shift is the small linguistic changes occur between ST dan TT. It means that a
grammatical unit in English, such as noun, affix, etc, has a lexical unit in Bahasa Indonesia as its
translation equivalent (Machali, 1998 : 14).
Based on the analysis of translation shifts on the received English email letters in to Indonesian
above, it is found that there are shifts of grammatical unit and structure concepts as well as linguistic
system changes from SL to TL as there are : (1) in level shift it was found that the grammatical
rule/pattern in SL changes in to lexical unit in SL without changing the original information of the SL, (2)
in structure shift it was found that phrase (modifier + noun) in SL change in to phrase (noun + modifier)
in TL and English phrase system change in to Indonesian phrase system without changing the original
information of the SL,(3) in class shift it was found that clause units in SL change in to word, phrase and
sentence in TL without changing the original information of the SL, (4) in unit shift it was found that
phrase unit in SL change in to word in TL and vice versa without changing the original information of the
SL, and (5) in intra-system shift it was found that plural noun/ noun phrase in SL change in to singular
noun/ noun phrase in TL without changing the original information of the TL.
The percentage of grammatical shifts of all data from received English email in to Indonesian
by google translation at PT. Mutiara Medica during December 2014 is shown in table 4.6 below :
Table 4.7 The Percentage of Grammatical Shifts Frequency in Received English Email
Letters in to Indonesian by Google Translation at PT.Mutiara Medica Sunggal During
December 2014
Types of Grammatical Data/f of Grammatical Shift
Translation Shift f f%
Shift A B C D E F
Level Shift Grammatical 0 4 8 7 3 2 24 10,53%
Rule
Lexicon
Phrase (m +n)- 10 18 16 12 14 6 76 33,33%
Phrase (n + m)
Sructure Shift Phrase -
Phrase 1 2 3 10 3 2 21 9,21%
(different
linguistic
System)
Clause Word 2 0 0 0 0 0 2 0,88%
Clause 0 0 0 0 3 0 3 1,32%
Class Shift Phrase
Clause 0 0 0 1 0 0 1 0,44%
Sentence
Phrase Word 3 5 4 5 7 3 27 11,84%
Unit Shift Word Phrase 4 9 9 8 13 8 51 22,37%
Plural
Intra System Noun/Noun 8 1 0 7 1 6 23 10.09%
Shift Phrase-
Singular
Noun/ Noun
Phrase
Total 228 100%

From the table above it is shown that from 228 data of translation shift in the 27
English email letters in to Indonesian by google translation taken from 6 different companies
during December 2014, it is found that the dominant grammatical shifts occure in the process of
translation is Phrase (Modifier + Noun) to Phrase (Noun + Modifier) which reaches 76 grammatical
shifts or it is equal to 33,33. The second level is word to phrase which reaches 51 grammatical
shifts or 22,37% of the overall data. The third level is phrase to word which reaches 27
grammatical shifts or 11,84 % , then followed by grammatical unit to lexicon, plural noun/ noun
phrase to singular noun/ noun phrase, pharse to phrase with different linguistic sysytem from ST
to TL where each of them reaches 24 shifts or 10,53 %, 23 shifts or equal to 10,09 % and 21
shifts or equal to 9,21%. The least levels are shifts of clause to phrase, clause to word and clause
to sentence which reach 3 shift or 1,32%, 2 shifts or equal to 0,88% and 1 shift or equal to 0,44%
of the overall data.

Research Findings
After the completions of the data analysis, some findings related to the reseach problems were
found that :
1. The types of translation shifts occur are due to the shifting of grammatical unit as well as the small
 linguistic changes occur between ST dan TT. It means that the shifting of grammatical unit and small
linguistic changes from SL to TL influce lexical unit in TL without losing the the original information
ot the SL.
2. The types of translation shift in the 27 English email letters in to Indonesian by google translation
taken from 6 different companies during December 2014 is stucture shift which reaches the highest
total frequency 97 shifts or 42,54% of the total 228 shifts. The second level is unit shift with 78
shifts or 34,21% of the total data then followed by level shift which reaches 24 shifts of the overall
data or 10,53%. The next level is Intra system shift which reaches 23 shifts of the overall data or
10,09%, and the last level is class shift which only reaches 6 shifts of the overall data or 2,63%. It
means that in conducting the process of translation google translation is more effective when it
translate in the level of word to word or phrase to phrase but not in the level of clause to clause or
sentence to sentence.
3. The dominant grammatical shifts occure in the process of translation is Phrase (Modifier + Noun) to
Phrase (Noun + Modifier) which reaches 76 grammatical shifts or it is equal to 33,33. The second
level is word to phrase which reaches 51 grammatical shifts or 22,37% of the overall data. The third
level is phrase to word which reaches 27 grammatical shifts or 11,84 % , then followed by
grammatical unit to lexicon, plural noun/ noun phrase to singular noun/ noun phrase, pharse to phrase
with different linguistic sysytem from ST to TL where each of them reaches 24 shifts or 10,53 %, 23
shifts or equal to 10,09 % and 21 shifts or equal to 9,21%. The least levels are shifts of clause to
phrase, clause to word and clause to sentence which reach 3 shift or 1,32%, 2 shifts or equal to 0,88%
and 1 shift or equal to 0,44% of the overall data. From this finding it is known that google translation
is effective when it used to conduct the translation from word to word, Phrase to Phrase, word to
phrase and phrase to word but not in phrase to clause , clause to clause or sentence to senctence.

Discussion
The above description and data analysis have shown that there are some points which can be
considered a the important ones to be discussed.
1. The occurances of translation shifts in the received English email letters in to Indonesian by google
translation at PT.Mutiara Medica Sunggal during December 2014 are due to the shifting of
grammatical unit as well as the small linguistic changes occur between ST dan TT.
2. The most dominant types of translation shift occured in the English email letters in to Indonesian by
google translation at PT Mutiara Media during December 2014 is stucture shift which reaches the
highest total frequency 97 shifts or 42,54% of the total 228 shifts. It means that in conducting the
process of translation google translation is more effective when it translate in the level of word to word
or phrase to phrase but not in the level of clause to clause or sentence to sentence, morever for the
complex sentence of an academic writing.
3. The most dominant grammatical shifts occured in the process of google translation is Phrase (Modifier
+ Noun) to Phrase (Noun + Modifier) , word to phrase and phrase to word . It means that google
translation is effective when it used to conduct the translation from word to word, Phrase to Phrase,
word to phrase and phrase to word but not in phrase to clause , clause to clause or sentence to
senctence. Because grammatical shifting is conducted in order to make the target language is more
natural and understandable by the readers.

5.Conclutions and suggestions

Conclusions
1. Based on the characteristics of each type of translation shifts observed in the analysis It can be
concluded that the types of translation shifts in the received English email letters in to Indonesian by
google translation at PT.Mutiara Medica Sunggal during December 2014 occured due to the shifting
of grammatical unit as well as the small linguistic changes occur between ST and TT. It is based on
Catford theory which stated that shifts is the replacement of one grammatical unit by another as well
 as the shifting of structure concept, The shifting of grammatical unit and small linguistic changes
from SL to TL influce lexical unit in TL without losing the the original information ot the SL.
2. The dominant type of translation shifts in English email letters in to Indonesian by google translation
during December 2014 is stucture shift which reaches the highest total frequency 97 shifts or
42,54% of the total 228 shifts. It means that in conducting the process of translation google translation
is more effective when it translate in the level of word to word or phrase to phrase but not in the level
of clause to clause or sentence to sentence, morever for the complex sentence of an academic writing,
because shifting is conducted in order to make the target language is more natural and understandable
by the readers
3. The dominant type of grammatical shifts occured in the process of translation of English Email letters
by google translation is Phrase (Modifier + Noun) to Phrase (Noun + Modifier) which reaches 76
grammatical shifts or it is equal to 33,33%..It can be concluded that google translation is more
effective when it used to conduct the translation from word to word, Phrase to Phrase, word to phrase
and phrase to word but not in phrase to clause , clause to clause or sentence to senctence especially for
the translation of formal academic writing. Because grammatical shifting is conducted in order to
make the target language is more natural and understandable by the readers.

Suggestions
Based on the findings, some suggestions are proposed as the following :
1. It is suggested to the English teachers who teach translation to give the students the knowledge of
translation shifts in conducting the process of translation in order to produce the good quality of
translation products as shifting in translation is to make the target language more natural and easier
understandable by the readers without losing the original information of the source language.
2. It is suggeted to the company, PT. Mutiara Medica Sunggal to use "Google Translation" only for
translating non academic simple sentences, it can still be used, but it must be accompanied by Human
Translator, who should edit and revise the post-translation result.The revisions should be done by the
human translator is in the revisions of grammatical unit related to different linguistics system between
the source languge (English) and target language (Indonesian).
In order to make a comparation of better result of translation products,it is also suggested to use
othe kinds of translation tools especially in translating the received English email letters.
3. It is suggested to other reseachers to use the catford theory of translation shift or other theories of
translation to analyze the mistakes found in google translation process as well as the application of
google translation on the other kinds of texts in conducting the process of translation in order to make
a comparison with this study.

References
Aiken, M., and Balan, 2011. An Analysis of Google Translate Accuracy. In Translation Journal,
Vol. 16, No. 2, April 2011 .Accessed December 4, 2014, from
http://translationjournal.net/journal// 51pondering.htm.

Aiken, M, Park, M., Simmons, L, and Lindblom, T. 2009. Automatic Translation in Multilingual
Electronic Meetings. In Translation Journal, Vol. 13, No. 3, July 2009 .Accessed on 1
December 15, 2014 from http://translationjournal.net/journal//49meeting.htm.

Arikunto, Suharsimi. 2006. Prosedur Penelitian Suatu Pendekatan Praktik.Jakarta : Rineka

Cipta.
Atherton, J.S. 2011. Teaching and Learning; Objectives Accessed on December 15,2014
fromhttp://www.learningandteaching.info/teaching/objectives.htm

Baker, M. 1992. In other words: a Coursebook on Translation. London: Routledge

Baker, M. 2000. In other words: a Coursebook on Translation. London: Beijing: Foreign Language
Teaching and Research Press.

Bassnett, S. 1998. Translation Studies: revisited edition. London: Routledge

Bell, R. T. 1991. Translation and Translating : Theory and Practice. New York: Longman
Brislin,R.W. 1990. Applied Cross Cultural Psychology (M). Newbury Park. California: Sage
Publications.

Brislin,R.W. 1976. W. Translation: aplication and Research. New York: Garduer Press..
Catford, J.C. 1965. A Linguistics Theory of Translation. London: Oxford University Press.

Catford, J.C. 1965 . A linguistict Theory Theory of Translation. Accessed in December 15, 2014 from
https://www.scribd.com/doc/97199603/95074003-J-C-Catford-A-Linguistic-Theory-of-
Translation-Oxford-Univ-Press-1965

Creswell, J. W.(2001).Research Design Second Edition. London: Sage Publications.

Echol & Shadily, 2010. An English Indonesian Dictionary: Cornel University

Fem, 2011. An Analysis of the Translation from English Sentences to Indonesian Sentences by Using
Google Translate Aaccessed in December 15, 2014. from
http:www//eprints.unika.ac.id/494/1/07.80.0019_Srestha_F.pdf

Fernald. J.C.1997. English Grammar Simplified. New York. Barnes:

Haliday, M.A.K, & Hasan, R. 1998. Language, Context and Text : Aspects of Language in a Social
Semiotic Prespective, Oxpord : Oxport University Press.

Hatim, B and Mason, I, 1997. Translator as Communicator. London and New York: Routledge

Hornby. AS. 1987. Oxford Advaced Learners Dictionary of Current English. London: Oxford University
Press.
Hutahuruk, T.R, 2013. The Translation of technical Text By Students of English Department FKIP
Universitas HKBP Nommensen Medan. Unpublished Thesis, Graduate School University HKBP
Nommensen: Medan.

Hutchins, J, 1995. Reflection on the History and Present State of Machine Translation.Accessed in
Decemebr 15, 2014 from http:// www.citeseerx.com .

Hutchins, J ,1999. The development and use of machine translation systems and computer-based
translation tools, 2-13. Accessed in December 15, 2014 from http:// www.citeseerx.com
Hutchins, J ,2002. Machine Translation: A Brief History. Accessed in December 15,2014 from
http://aymara.org/biblio/mtranslation.pdf

Isabelle Ee Hsia, T.,and Tang Enya K ,2004. A Discourse of Cognitive Paradigm

on the Accuracy of Machine Translation. Accessed in December 15, 2014 from http://
www.citeseerx.com

Kawulich. B, 2004. Data Analysis Techniques in Qualitative Research. Accessed in January 15, 2015
from http://www.eeraonline.org/JRE-2004-08.
Martin. J.R, 1992. English Text : System and Structure. Sydney University.
Miremadi,A. 1995. Theory of Translation and Interpretation. Tehran:SAMT
Miles,M.B. & Huberman, A.M. 1984. Qualitative Data Analysis ; a Sourcebook of New Method.
California : sage Publication
Muhtalifah. 2011. An Analysis of English Indonesian Translation Shift in the Hannah Banana Short Story
By Ribut Wahyudi. Accessed in December 15, 2014 from
www.http://repository.uinjkt.ac.id/dspace/bitstream/12345678/4857/1/SITI%20MUHTALIFAH-
FAH.PDF

Munday, J. 2001. Introducing Translation Studies: Theories and Application. London and New York:
Routledge

Newmark, P. 1991. Approaches to Translation. Oxford: Pergamon Press. Ltd

Newmark,P. 1988, A Textbook of Translation,Prentice HallInternational (UK).