The different bibles in common language are all a translation of the original Greek text. Where to find this Greek text is a complex question.

1. Manuscript Evidence for the Text

   There are about 3,000 manuscripts (complete or partial) of the Greek NT copied between the 2nd and 17th centuries, to which must be added 2,200 handwritten lectionaries organized according to the liturgy of the church from the 7th century onwards. All these texts do not always agree with each other, but these differences are rarely significant. No original or autograph manuscripts have been preserved. The differences are either the result of a copyist's errors or a deliberate change, because the copyist wanted to improve the Greek text, or to modernize the vocabulary, or to introduce an explanation, or to harmonize with another passage, or to omit something that seemed doubtful. It is not because one manuscript is very old that it is better than another, for a 6th century text may be the only remaining copy of an older but lost copy, which was closer to the autograph than a 2nd century copy.

   1. Textual Families

      Biblical scholars have organized manuscripts that share similar features into groups or families, but none of these groupings of texts has uncontaminated descent from the originals. The most commonly recognized are the following:

      1. Alexandrian

         By the end of the second century, Christian scholarship was flourishing in Alexandria, and over the next few centuries manuscripts were carefully copied by scribes who had a sophisticated appreciation of Greek, to the point where one wonders if they did not improve the text. It is characterized by shorter readings.

      2. Western

         This is a catch-all group, named after the western circulation (North Africa, Italy, Gaul) of some of the Greek manuscripts. However, alongside them are Greek manuscripts that can be linked to Egypt and the East Syriac-speaking churches. Often the texts are a paraphrase, so that they are longer than the lean and spare Alexandrian readings, as if words had been added (interpolations); but in a significant number of cases for Luke's gospel, the Western text omits what is in the Alexandrian. In Acts, the text is almost a tenth as long as that found in the Alexandrian tradition.

      3. Caesarean

         In the third and fourth centuries, Caesarea on the Mediterranean coast was the most important Christian center in Palestine, and boasted a large library and the biblical scholars who used it. The basic text of this group, dating from the beginning of the third century, was probably brought there from Egypt; it then spread to Jerusalem and, through Armenian missionaries, to Georgia in the Caucasus. In its development, the Caesarean textual tradition is situated between the Alexandrian and the Western tradition.

      4. Byzantine (or Koinē)

         This text, which is in fact a combination of texts smoothing out difficulties and harmonizing differences, was used in the liturgy of the Byzantine church (becoming almost normative from the 6th century onwards), and is generally considered a rather late and secondary development. However, some of its readings are ancient and date back to the church of Antioch around the year 300.

   2. Textual Witnesses

      Among biblical scholars, there are three types of textual witnesses

      1. Papyri (abbreviated P)

         The oldest documents on papyrus have been discovered in Egypt. Since 1890, about a hundred have been discovered, which paleography dates from the 2nd to the 8th century. Among the oldest, let us name:

         • P⁵²: consisting of a fragment smaller than an index card on which is inscribed part of John 18:31-34. Its date of about 135 makes impossible theories of a very late dating of John.
         • P⁴⁶: dated around 200 or earlier, it contains the Pauline epistles, including Hebrews (following Romans, an order based on decreasing length), but not the Pastorals. It belongs to the Caesarean group.
         • P⁶⁶: dated around the year 200, it contains a heavily corrected text of a large part of John. It belongs to the Alexandrian group, close to the text of the Codex Sinaiticus
         • P⁷⁵: dated 225, it contains Luke 2:18 - 18:18 and Luke 22:4 as well as John 15:8. It belongs to the Alexandrian group, close to the Codex Vaticanus.

      2. Great Uncial Codices

         The codex or book is made up of sheets of parchment on which one wrote in print (capital letters). It reached its peak between the 3rd and 9th centuries. The legalization of Christianity in the 4th century under Constantine made possible the public existence of centers of learning and monasteries where numerous codices were copied and preserved. They often contain the entire Greek Bible and sometimes some of the earliest non-canonical Christian works. Of the approximately 300 known uncial codices, the most important (designated by capital letters), starting with the oldest, are:

         • B (Codex Vaticanus), mid-4th century, which lacks the last part of the NT. It illustrates the type of Alexandrian text and is considered by most scholars to be the best witness to the original NT text.
         • S ou א (Codex Sinaiticus), mid-4th century, containing the entire NT plus The Epistle of Barnabas and the Pastor of Hermas. It follows the Alexandrian tradition in the Gospel and Acts, though elsewhere it presents Western readings.
         • A (Codex Alexandrinus), early 5th century, once contained the entire NT plus 1 and 2 Clement and the Psalms of Solomon; unfortunately, pages have been lost. In the Gospels, the text is Byzantine, but Alexandrian in the rest of the NT.
         • D (Codex Bezae), 5th century, contains Matthew, John, Luke, Mark, 3 John and the Acts in Latin and Greek on facing pages. It is the main representative of the Western textual tradition.

      3. Minuscules

         By the ninth century, a cursive style of writing (bound letters) began to supplant the uncial texts, and there are nearly 2,900 NT manuscripts in this script. Two families of these, named f1 and f13 (after biblical scholars K. Lake and W. H. Ferrar) are witnesses to the Caesarean text tradition.

      In addition to all these witnesses, we can add translations in other ancient languages, because they testify to a Greek text that served as a basis for the translation. Around the year 200 translations were made into Latin and Syriac, called "old Latin" and "old Syriac" to distinguish them respectively from the Latin translation of the Vulgate by St. Jerome at the end of the fourth century, and the Syriac translation of the fourth to fifth centuries (the Peshitta), which became the standard Bible of the Syriac Church. These old translations belong to the Western textual tradition. Finally, the Church Fathers also provide information on the Greek text or translation that was submitted to them.

2. Observations about the Use of the Evidence
   • Many of the differences in the biblical text found in the great uncial codices of the fourth and fifth centuries already existed around the year 200, as papyri and early translations show. How could so many differences appear less than a hundred years after the writing of the original texts? Perhaps the answer lies in the attitude of the copyists towards the NT books they were copying. For them, there was no slavish devotion to their exact wording. So they did not hesitate to comment and interpret what they copied. It was later, when more settled ideas about canon and inspiration shaped the mindset, that the focus shifted to maintaining the exact wording.

   • Sometimes, when faced with competing readings, a decision cannot be made simply on the basis of the manuscripts, because the weight of the textual witnesses is equally divided. We must then ask why a copyist would have changed a text, and this question can give us insight into his theology. For example, some Western manuscripts do not contain Jesus' words from Luke 23:34a, "Father, forgive them, for they know not what they do." Did a devout copyist add this phrase to the original Lucan text, which lacked it, because he thought that Jesus would surely have thought that way? Or did a copyist remove it from the original because the Church Fathers taught that those who put the Son of God to death could not be forgiven?

   • The Reformation gave a major impetus to obtaining vernacular translations of the original Greek because they would be more accurate and accessible to people than the Latin Vulgate. Thus the 1611 version of the standard English translation of the King James was made on the basis of Erasmus' edition of the Greek NT designated as the accepted or received Greek NT, the Textus Receptus. Unfortunately, Erasmus relied heavily on 12th and 13th century manuscripts from the Byzantine tradition; he had no papyri available and did not use any of the great codices listed earlier. Ironically, the Latin Vulgate, translated 1100 years earlier, was based on better Greek manuscripts.

   • In the late 19th century, scholarship finally won the battle to replace the inferior Textus Receptus with new editions of the Greek NT based on the great uncial codices and other textual witnesses available since the time of Erasmus, and these editions have since undergone corrections in light of new discoveries. The best known among students is the Nestle-Aland edition (constantly updated), which is also used in the United Bible Societies' edition of the Greek NT. But it should not be forgotten that this is an eclectic text, drawing on one tradition for one verse and another tradition for another verse, a text that never existed as a unit in antiquity and was never read in any Christian community. A corollary is that if the books of the NT are canonical, no particular Greek text should be canonized; and the most that can be claimed for a critically prepared Greek NT is acceptance by biblical scholars.

   • The Roman Catholic Church decided on canonicity on the basis of long-standing regular use in the liturgy, not on the basis of biblical scholars' judgments about the author or copyist. Thus, the story of the woman caught in adultery in John 7:53 - 8:11 and the long ending of Mark (16:9 - 20) were designated by the Council of Trent as belonging to Scripture, even though they are absent from many NT textual witnesses. Catholics remain free to accept the judgment of competent biblical scholars that these passages were not part of the original text of the respective gospels.

   • Textual criticism remains a difficult exercise. Recent translations of the NT sometimes include footnotes to readings that differ from those chosen in the text. To discover the interesting side of textual criticism, here are some examples where we can verify the choice made by the translator of the NT text we have in our hands:
     1. John 1:18: "God the only begotten Son" or "He is the only begotten Son"; in the first case, the son is called: God
     2. Luke 24:12, describing Peter's race to the tomb of Jesus, is missing from some witnesses. If this version is original, it constitutes a very close link between Luke and John, the only other Gospel where Peter runs to the tomb (20:3-10).
     3. Eph 1:1: Some witnesses do not refer to Ephesus. One suggestion is that this was a general missive with a blank space that could be filled in with the name of the congregation to which it was read.
     4. John 7:53 - 8:11, the story of Jesus forgiving the adulteress is absent from many manuscripts and, according to many scholars, was probably inserted into John long after the original Gospel was completed. However, other scholars believe that this story was removed from the original text because it ran counter to the early Christian practice of denying public forgiveness to adulterers?
     5. Mark 16:9-20 and two other alternative endings were apparently added by the scribes to soften the abruptness of the Gospel's ending with 16:8 (where the women said nothing to anyone). What are the implications if Mark originally ended without describing a post-resurrection appearance? Since Mark 16:9-20 is found in most Bibles, what is the impact when 16:1-20 is read sequentially?