The Great Bible Review

The review method

I have chosen a very limited set of only 8 verses for this review, with a preference for quite short verses. I wanted to optimize the work load, knowing that each verse needs to be reviewed hundreds of times. One of the things that I learned from the aforementioned Bible review project is to keep it simple and minimalistic, or else your research will take decades to complete. This review project has taken hundreds of working hours, during approximately half a year. Without paying attention to optimization and simplification of the workload, it could have taken thousands of working hours, and half a decade.

Many of the test verses include scenarios where the translators may feel tempted to use a paraphrase, or “helpfully” tell the reader something that the original Bible text does not say. The name of God and the name of the Messiah are included in the test verses. Some readers consider it crucially important, how these are spelled in an english Bible. Psalms 110:1 shows how clearly the translations differentiate the proper noun Yahweh from the common noun Ȧdonī. Zekhariah 8:18 contains the expression źėvạōt in the context of God, which is often left untranslated as something like “Sebaoth”. Markos 12:35 shows how the translators handle a complex and inconvenient sequence of verb forms, and Romans 8:12 serves as an indicator of gender-neutral translation principles. Loukas 3:4 demonstrates some additional material that the Latter-Day Saints have in their Bibles. None of the test verses indicate controversial doctrinal preferences of the Jehovah’s Witnesses.

The method that I use for reviewing Bible translations is my own design. It differs from the method used by the aforementioned group of private Bible researchers, partly because of my different personal preferences, and also to minimize the work load. I have heard rumours that some organizations such as Gideons International probably review the accuracy and theological orthodoxy of Bibles, when they choose which translation to distribute in each language. I have no knowledge about their review methods, however.

Many private persons have published diagrams on the internet, which position some of the most famous Bible translations on a scale from strictly literal to freely rewritten paraphrase. I have contacted several such persons, to find out how their reviewing method and research data looks like, but most of them have admitted that their diagrams are based on a very rough and subjective estimation: a “feeling” rather than systematic research. Dr. Andi Wu of the Global Bible Initiative (formerly known as Asia Bible Society) has analyzed 9 Bible translations with a computer algorithm, which intends to avoid the problem of subjective decisions taken by a human, but which also fails to indicate many relevant aspects that my handmade research indicates.

Theological orthodoxy is something that my review does not attempt to assess. I give an extra high deviation score of 9 points for deliberately mixing up the main theological concepts, though (e.g. replacing Yahweh with Jesus in OT, or replacing Jesus with Yahweh in NT). Reprehensible translation choices can exist in isolated verses anywhere in the Bible, so assessing the theological orthodoxy of a translation would require reading the whole Bible through. Such an effort is beyond the scope of this project.

I have not assessed or documented, which manuscripts in the original languages each translation uses. Or from which language the translation was done: from hebrew, greek, aramaic, latin, english or french. Finding such information would be difficult or impossible for many of these translations. There are other websites that specialize in assessing the theological orthodoxy or the manuscript choices of Bible translations.

If a translation choice is grammatically possible and logically sensible, my review accepts it without deviation points. Words and expressions can often have several possible meanings, and I cannot claim to know, which of the possible meanings was intended by the original writer. If the Jehovah’s Witnesses translate a phrase in a very unusual way, by adding a comma to split a sentence into two separate thoughts, my review would not indicate that anything wrong has been done, if the interpretation is grammatically possible. This belongs to the field of theological review, not grammatical review. The original manuscripts don’t have punctuation, so it is not possible to prove where the thoughts are intended to be continuous and where separated. This aspect is relevant in three out of four New Testament verses in this review: Mattityah 19:12, Loukas 3:4 and Romans 8:12.

I explain the most significant decisions and policies of the review in the introduction of each verse. Otherwise I give very sparingly any explanations for the deviation markings. If I had done so, it could have multiplied the workload of creating this review. Reading the review would also be a heavier effort. I provide a detailed description of the general review methods, which should explain nearly all scenarios in the deviation markings.

I have analyzed the Bible translations concerning these four aspects: literal accuracy, form-faithfulness, correctness of the main thoughts, and convenient readability for the modern mainstream population. Separate scores are given for these aspects, so the reader can choose, which of these indicators are most relevant for assessing the quality of Bible translations.

I review the main text of each Bible translation, without footnotes and without references to other Bible verses. The translated text is compared to widely accepted versions of the hebrew and greek manuscripts. This concept is not always very precise, because there are some variations in manuscripts, but in the reviewed 8 test verses the differences in manuscripts have very little importance. It is more significant if a translation of the Tanakh (Old Testament) is based on the greek Septuagint, or if a translation of the New Testament is based on the aramaic Peshitta. Also these translations are compared to the same hebrew or greek manuscripts as any other translation.

As the “Bible manuscripts” I present to the reader the text WLCR for verses of the Tanakh, and BYZR for verses of the New Testament. These texts are displayed in the latin alphabet, using my romanization standards HE-RZ-P and GC-RZ-P.

GC-RZ-P is very similar to the romanization of greek used by biblehub.com and other popular Bible websites for english-speakers. The main difference is that I indicate the grammatical iota as a => ă, ē => ĕ and ō => ŏ, and I write G in certain situations as Ñ, e.g. añgelos. (The romanization of biblehub.com fails to indicate the grammatical iota, and carelessly uses an ordinary N for G in words such as añgelos.) I omit nearly all traditional accent marks, to reduce unnecessary information noise. Instead I use accent marks for a purpose that can be useful for the modern reader: to indicate that a word has a meaning other than its most common meaning, e.g. kai = “and”, but kài = “also”. Examples of such spelling variants are listed in this document.

I reduce the unnecessary hellenization of hebrew names in BYZR, compared to traditional diacritics that are used in greek manuscripts, by writing a strong breath “Ḥ” between vowels in names such as Isaàk => Isaḥak, if the original hebrew name has a Hey or Ḫet between the vowels. (The dot under letter H does not indicate a hebrew Ḫet sound, but rather that the strong breath in greek is untraditional, and does not belong to the traditional ortography of this word. Using a special character Ḥ instead of an ordinary H in this scenario makes it easy to automatically remove such untraditional ortography from the proper nouns, if the reader prefers to see proper nouns written in the traditional style of koiné greek.)

I also use “J” instead of “I” in BYZR in the first letter of hebrew names such as Iēsous => Jēsous, which begin with the consonantal Yod sound in hebrew — to suggest a pronunciation Yē- rather than Iē- (but not Dzē-), if the second letter in the greek name version is a vowel. (The aforementioned name Isaḥak does not meet these criteria, because the second letter is not a vowel.) The standard pronunciation of koiné greek would apparently never begin a word with a Yod sound, but many people in the modern world may feel that the pronunciation Iē- is inconvenient and artificial, and the pronunciation Yē- is easier and more beautiful, no matter if it is historically correct or not. The authors of the greek manuscripts of the New Testament may have spoken greek as a foreign language (though mastering it well), possibly with an accent that may have included some hebraicisms, such as pronouncing the I as Yod at the start of many hebrew proper nouns. The untraditional letter J can easily be automatically converted into traditional ortography I, because the letter J has no other purpose in this transliteration standard than marking the letter I.

I have a tendency to split long hebrew proper nouns into two separate words, while koiné greek texts traditionally write them together: e.g. Bēthsaida => Bēth Saida, or Kapernaoum => Kaper Naḥoum.

My romanization standard HE-RZ-P for hebrew differs quite much from any other transliteration standard, and documentation is necessary for interpreting it correctly in some scenarios, especially such where consonants àyin or ạlef are involved. (I will publish the documentation of this transliteration standard probably during the spring of 2024.) I may sometimes use untypical ortography in WLCR and WLCRA to indicate that a word has a meaning other than its most common meaning, e.g. in Other Laws 23:10 kiy = “if”, but in most other circumstances kī = “for / because”. (According to the standard HE-RZ-P, “ī” and “iy” are synonyms, which produce the same outcome in hebrew script.)

The text BYZR is Revised Byzantine Textform, which is based on the Byzantine Textform 2018 text, with my punctuation and ortographic choices, which have been discussed afore. WLCR is Revised Westminster Leningrad Codex, based on the UXLC 2.0 2023 text, which contains the original consonants, my punctuation, and vowelization that is based on the masoretic tradition, with my minor modifications, which are stylistic and do not affect the meaning of words. The main differences are these: 1) I use uppercase and lowercase letters, according to modern conventions. 2) I write prepositions and the word “and” as separate words, according to modern conventions. If the preposition should cause the first letter of the next word to be doubled, this is not marked in writing: e.g. bammidbạr => ba midbạr. 3) I replace all ceremonial pausal forms with ordinary forms. 4) The ortography of a word never changes because of its position in sentence. I don’t soften the first consonant of a word because of a preposition, e.g. “le Veyt Leḫem” => “le Beyt Leḫem”. I leave the softening of consonants after a preposition as a matter of pronunciation, which is not marked in writing. 5) In the name of God I use the most widely accepted (though still uncertain) vowelization Yahweh, instead of the six different intentionally misspelled variants used by the masoretes, the most common of which is Yėhowạh.

In addition to the “original” Bible text, I present also an amplified version WLCRA of the hebrew Tanakh text, which contains added words that help the reader to understand difficult or ambiguous parts of the Bible text. The added words are marked with italics or square brackets. I present also two translation samples, SAMPLET and SAMPLEP, which I have prepared myself. SAMPLET demonstrates how such a translation might look like, which gets a clean score in this review for all other aspects except fluent readability. (Some famous literal translations get a nearly clean score in this review, without even knowing the rules by which they will be measured.) SAMPLEP demonstrates how a fluently readable and easily understandable paraphrase might look like, without making heavy compromises in literal accuracy or correctness of the main thoughts. Only some verses exist of the texts WLCR, WLCRA, BYZR, SAMPLET and SAMPLEP: they are not complete versions of the Bible.

The text of each Bible translation is presented in this review with nearly identical text styles as the translation itself uses for each word (e.g. plain text, italics or bold). However, the colour of text is not replicated or taken into account in the review. Brackets are used nearly identically: only very unusual brackets are replaced with more typical ones. Any footnotes and references to other Bible verses are omitted, but other comments within the verse text are included in this review.

Line breaks are replaced with spaces. If a translation uses line breaks in the role of punctuation (where one would expect a comma), and the text would feel inconvenient to read without the line breaks, I use slashes (/) where the translation has line breaks. Slashes in medieval texts are transcribed as commas, except in early Wycliffe Bibles and WYE1848, which use slash (/) in the role of the modern period (.) or colon (:), and dot at a varying height (. or ·) in the role of comma (,), and colon (:) in the role of comma (,) or semicolon (;), and tricolon (⁝) in the role of the modern semicolon (;) or colon (:). I transcribe these as / => . | · => , | : => ; | ⁝ => :. WYE139B sometimes uses a combined character of S and tilde, when one would expect a high dot (“comma”) to follow the word. Apparently this is a ligature of letter S and a high dot. I have transcribed these tildes with a comma, similarly as I would transcribe a high dot.

A deviation in form-faithfulness usually receives 1 penalty point. Other types of deviations usually receive 1, 2 or 3 penalty points. Higher scores of 4 - 9 points are possible, but extremely rare. 3 deviation points is the standard score for a strongly changed meaning, 2 points for moderately changed meaning, and 1 point for slightly changed meaning. 9 points is the standard score for mixing up the main theological concepts, such as replacing the name Yahweh with the name Jesus.

My decisions of giving 1, 2 or 3 penalty points to a deviation may contain some subjectivity. Also the scores for convenient readability and for the correctness of the main thoughts may contain subjectivity. A different editor might have made different decisions in many cases.

1) Literal accuracy

This deviation score indicates that a significant main word in the translated text has a different meaning than the word in the original language. Prepositions and some other small words are not included in this score, such as “to”, “in”, “under” or “over”. They are handled in the review of form-faithfulness. (In hebrew and greek, such details of grammatical meaning are often expressed with a suffix in the main word, while english language uses prepositions that are separate words.)

Added main words usually receive 3 deviation points for literal accuracy, and less significant words receive 1 deviation point. (Added main words that repeat a thought, which is expressed in any case, receive a lenient deviation score of 1 point.) I have allowed without deviation points such added words, which are marked as not part of the original Bible text, e.g. with italics. Some Bible translations make mistakes in the use of italics, so that a word is marked with italics, although it is found in the original Bible text. These cases are handled harshly as missing words, because the translation gives a misleading impression to the reader that this word is not in the original Bible text. In the Tanakh (Old Testament) a common reason for such a “mistake” is that the translation follows the greek Septuagint, and marks with italics words that are in the hebrew manuscripts but not in Septuagint.

Some translations use italics for a different purpose than marking added words, for example to highlight proper nouns or important theological concepts. Some translations use square brackets or ordinary brackets for marking added words. I take flexibly into account the practices of each translation, without giving penalty points for unclarity in these markings, such as using ordinary brackets, which the average reader might not interpret as markers of words that are not part of the actual Bible text.

Some Bibles translate into english all proper nouns, such as the names of people. I have deemed this acceptable without deviation points, even if the result is formally different than the original text. (Usually the proper noun is very compact, but its translation is a complex paraphrase.) In the context of Bibles the attitude is understandable that the reader should be given a chance to understand the meaning of proper nouns. I have not rated the accuracy of such translations of proper nouns: they are always marked with 0 deviation points, regardless of how accurate the translation is. It is my general policy that I don’t want to punish a translator for making mistakes in an effort, which the others have not even attempted to do.

Bible versions write proper nouns in many different styles. The review of fluent readability assesses how convenient the style of proper nouns is to read (for the english-speaking mainstream population), but the other reviewed aspects do not pay much attention to the style of proper nouns. The Bible itself treats foreign proper nouns quite nonchalantly. Non-hebrew names are written in an approximate form in the Tanakh (Old Testament), which may differ quite much from the way how the name was written and pronounced in its original language. The society of Israel was multi-lingual in the New Testament era, and a person might use three or four different versions of his or her name, depending on which language was spoken in each situation.

The greek manuscripts of the New Testament translate the name of God into greek as Kyrios, which means Lord. Because of the uncertain vowelization and meaning of the name of God in hebrew, I have accepted without deviation points a wide range of translated meanings and written forms. I have given deviation points for two reasons only: 1) If this name can be mixed up with other concepts, because it is not written in a unique way. For example, some Bibles write Yahweh and Ȧdōnạy identically as “Lord”. 2) If the name Yahweh is deliberately replaced with a different theological concept, such as Jesus.

Many people have a strong preference that the name of God or the name of the Messiah must be written exactly in the form that they prefer. This review can be useful for finding and comparing Bible translations with the preferred form of these names, as the review shows how these names are written in 500 Bible translations, in Psalms 110:1, Zekharyah 8:18 and Markos 12:35.

Some Bibles, whose target audience is english-speaking jews who understand some hebrew, leave many words untranslated in the hebrew Tanakh (Old Testament), using them as loan words. The average english-speaker would not understand those words, but the target audience of such a translation is assumed to understand them. This Bible review intends to assess, how precisely the translators have converted the text into another language. Therefore I have given one deviation point for leaving hebrew common nouns untranslated, because it remains unproven, what the translators have thought these words to mean (or what the english-speaking readers understand them to mean). Deviation points given for this reason can seem unnecessary and unfounded, but for the sake of fairness I didn’t want to give a perfect and clean score for those who have not even tried to translate a word, when others are punished for every small mistake that they make while trying to translate the word into english.

Zekharyah 8:18 contains the hebrew expression Yahweh-źėvạōt, which means Lord of multitudes. If the word źėvạōt is left untranslated, as something like “Sebaoth”, I have treated it strictly with 3 deviation points for literal accuracy, because it makes this common noun look like a proper noun, and the average reader does not know what “Sebaoth” means, or that it should mean anything at all. In this scenario I give a deviation point also for form-faithfulness, because the expression Yahweh-źėvạōt contains a genitive in hebrew, but the expression “Lord Sebaoth” is not understood as containing a genitive in english language. I give a deviation point also for the lack of plural, because the word źėvạōt is plural in hebrew, but nearly all english-speaking readers understand the word Sebaoth as singular. These policies may seem harsh regarding the ethnic translations for english-speaking jews, whose readers are assumed to understand the meaning and grammatical aspects of such untranslated words.

When hebrew (or arabic) loan words are used in verses of the New Testament, I have accepted this with symbolic 0 deviation points, because the translators have correctly converted the word into another language — though not into english. In some ethnic translations this creates the situation that hebrew loan words are accepted with 0 deviation points in the New Testament, but punished with 1 deviation point in the hebrew Tanakh (Old Testament). My intention is to evaluate the efforts and performance of translators equally, which has led to this policy. A different intention would lead to a different policy. This aspect affects only a few ethnic translations, none of which would be in top ten of the most accurate Bibles in any case.

2) Form-faithfulness

This deviation score indicates that the translation uses a different form of a word than the source text. For example, a verb in the past tense is changed into future tense, or a noun is changed into an adjective, or the preposition “in” is changed into “out”. Deviations in form-faithfulness (and sometimes also in literal accuracy) are approved with 0 penalty points, if a formally similar expression is not possible in english, due to differences in the grammars of the languages. Use of the definite article or indefinite article is not assessed at all in this review, because the biblical languages do not have an indefinite article, and they use the definite article very irregularly, and in different scenarios than modern english.

Every missing or added word has generally received one deviation point for form-faithfulness. However, if a preposition and a main word are added into the text, the added preposition does not get a deviation point for form-faithfulness, because it defines the form of the main word, but the added main word does not have a “correct” form, to which this form could be compared. Hebrew and greek texts use a separately written personal pronoun with a verb less commonly than is customary and even necessary in english. I have deemed it acceptable without deviation points, if a separately written personal pronoun is added to a verb that contains a reference to that person, whenever the translators want to do so.

If an untypical and inconvenient verb form, such as “saying”, is translated with the expression “and said”, to improve the readability and style of the text, I treat it leniently with only one deviation point for form-faithfulness, which is marked on the verb, and the added word “and” is marked with 0 deviation points for literal accuracy and form-faithfulness. If a translation uses the expression “there is”, while the original text does not have the word “there”, the idiomatic addition “there” in english is marked with 0 deviation points for literal accuracy and form-faithfulness. This is a typical and expected expression in english language, in which the word “there” refers simply to existence, not to any specific location. The most conscientious translations use italics in the word “there”.

The verb “to be” in the present tense is often left unwritten in hebrew and greek language. Therefore we cannot be certain, when and where the original writers have intended to include the meaning “to be”. Deviation points are never given for an added verb “to be”, if such an interpretation of the text is grammatically possible.

It is also sometimes uncertain, how the original Bible text was intended to be divided into phrases, and sometimes even words. In rare cases punctuation can be a reason for deviation points, if it causes the meaning of the text to differ from any grammatically possible interpretation of the original Bible text. This review would permit breaking the hebrew or greek text into words differently than is the traditional interpretation, or using different than the traditional vowelization in some hebrew words, if it results in a credible theory of what the original writer may have intended to say.

3) Correctness of the main thoughts

This deviation score indicates that the translation does not express precisely the most important main thoughts of the verse, or new thoughts have been added to the text. Deviation points are not given for added thoughts that are marked as not part of the original Bible text (e.g. with italics or square brackets). A lenient deviation score of 1 point is given for thoughts that are included in the original Bible text, but misleadingly marked with italics or square brackets in the translation.

Which thoughts did the original writer intend as the most important message of the text, and which details are intended to be less important? It is impossible to give an authoritative answer to this question. I have selected the main thoughts according to my own interpretation and understanding.

4) Convenient readability

This score indicates that the text of a translation is not very fluent to read or easy to understand for the modern mainstream population, which is not accustomed to archaic language, rare slang words, or uncommon theological terminology. I have given a varying number of penalty points for archaic or rare words, and for leaving common nouns untranslated. Someone with strong reading skills and a vast vocabulary may think that I have given penalty points for many words that are not rare or difficult. It should be noted that large parts of the population have weak reading skills and a limited vocabulary, and the situation is not getting any better in future. I have reagarded it as possibly useful to point out, if a Bible translation uses vocabulary that is more difficult than would have been necessary.

I have required that proper nouns must contain readable consonants and vowels, and the text must be written in latin alphabet. Therefore the unvowelized word YHWH has received penalty points for difficult readability, from the point of view of the mainstream population. New loan words from hebrew or other languages have been rated as unintelligible for the average reader of english, although they might be understood by the ethnic target group of the translation that uses such words. However, if a rare or difficult word is explained immediately after the word (e.g. in parentheses), in that case the score for inconvenient readability is lenient, one point only (because of the trouble of reading the additional explanation).

Cumbersome or illogical use of punctuation has also received penalty points for inconvenient readability. If the translation contains additional comments or variant readings in parentheses or square brackets, each text block in parentheses has received one deviation point for inconvenient readability, because the additional comment breaks the fluent rhythm of reading the main text.

Bible translations often fail to define clearly with punctuation (e.g. with quotation marks), where a quotation begins and where it ends. This aspect of convenient readability has been ignored in the review, because this is a quite common practice in english language, and because the original Bible text does not indicate it either, where a quotation ends. Some translations may intentionally leave it ambiguous, where a quotation ends, to avoid interpreting the text on behalf of the reader.

I have rated also the original Bible text in hebrew or greek for convenient readability. Text of the hebrew Tanakh is rated with the assumption that the reader masters modern hebrew, but is not accustomed to the archaic grammar and vocabulary of the Tanakh. Text of the greek New Testament is rated with the (quite abstract) assumption that the reader masters koiné greek, but is not familiar with specialized theological terminology.

A detailed description of the review process and deviation markings is given on the page Deviation scores.