Father Pádraig McCarthy, a retired priest of Dublin, Ireland, has sent to us the following data concerning the readability of the new translation of the four Eucharistic Prayers. We are pleased to offer this data to our readers.

We have also included some of Father McCarthy’s comments, offering more background and discussing the import of these findings, which we elicited in a follow-up e-mail exchange. They appear below, following the test results.

Our thanks to Father McCarthy for this contribution to Pray Tell. – Editor

  • The Eucharistic Prayer texts were tested without acclamations, prefaces or doxology.
  • These tests do not involve any theological or liturgical positions. The results have serious implications for the suitability of the texts.
  • “Readability” refers to reading text from the written page.
  • A person reading from the page can go back to the beginning of a long or difficult sentence to check where the meaning comes from. A person listening cannot do such a second take.
  • The comprehensibility of the text as delivered in speech will depend on the ability of the person speaking.
  • In the new translation, the longest single sentence in the four Eucharistic Prayers is the “Communicantes” of the first Eucharistic Prayer: “In union with the whole church …”. This has 82 words, but, on account of the list of names, is a little more acceptable. In the current translation, it is in three sentences, but four effectively, with a semi-colon break. The second longest sentence is the opening sentence of the Third Eucharistic Prayer, with 72 words. In the current translation, this is three sentences, with 61 words in total.
  • The current translation of the First Eucharistic Prayer (Roman Canon) has 18 sentences before the words of consecration. The new translation reduces this to 8 sentences. All but one are over 40 words long. (The sentence structure of two of those would allow them to be spoken as two units.) An occasional long sentence is not usually a problem. A succession of them most definitely is.

The results of testing the four Eucharistic prayers in the current and forthcoming translation are as follows:

New – 2010
Number of characters (without spaces) 10,818 9,046
Number of words 2,549 2,177
Number of sentences 72 106
Average number of characters per word 4.24 4.16
Average number of syllables per word 1.46 1.42
Average number of words per sentence
35.40 20.54
Gunning Fog* index 17.21 10.75
Coleman Liau** index 8.35 7.21
Flesch Kincaid grade level 15.40 9.14
ARI (Automated readability index) 16.26 8.41
SMOG 13.10 10.20

* Gunning Fog index: Indication of the number of years of formal education that a person requires in order to easily understand the text on the first reading

** Coleman Liau index: Approximate representation of the U.S. grade level needed to comprehend the text

Change in statistics from current translation to new

2549 – 2177 = 372 increase (17.09% increase)

106 – 72        = 34 decrease (32.08% decrease)

Average words per sentence

35.4 – 20.54 = 14.86 increase (72.35% increase)

Flesh Kincaid Grade level :

15.4 – 9.14    = 6.26 increase (68.49% increase: Increase from 9.14 years of education to 15.4 years.)


The grids give us data according to a number of different assessment tools. We could make it simpler by looking just at the statistics for the average number of words per sentence: in this case, going from 20.54 for the current translation to 35.4 for the new translation.

This alone is startling. I consider it very serious, in that it will make full active conscious participation in the Liturgy very difficult for many people who do not have a college education, and even for many who have! It goes contrary to what paragraph 25 of Liturgiam Authenticam states: “So that the content of the original texts may be evident and comprehensible even to the faithful who lack any special intellectual formation, the translations should be characterized by a kind of language which is easily understandable.” I see it as a matter of justice if we are to be followers of the one who came especially for the most vulnerable in society.

It may be useful to provide some comparisons. Barack Obama’s speech at his inauguration has average 20.42 words per sentence. He is an effective and dignified communicator. The legal article by Bishop Brian Dunn which I found on your website (thank you!) comes in at 19.62. Pope Benedict’s letter for World Day of Peace 2011 reaches 32.14 – the highest I’ve found, apart from the new translation of the Eucharistic Prayers; and his letter is not, I think, designed with a view to reading aloud in public.

A wide variety of tools assess readability – there are perhaps 200 different systems, some for specific purposes; for instance, to assess the suitability of a book for a particular grade level at school, or technical manuals. (Think how difficult it can be to understand the User Manual for a new piece of office or kitchen equipment!) They work on a number of factors, depending on the specific application. Factors usually include the average number of words per sentence and the number of polysyllabic words; some have a list of words in everyday use, and then check the number of unusual words of whatever length. The key factor here is always to compare like with like – not two different tools.

The best known one seems to be the Flesch-Kincaid Grade Level, which has gone through a number of revisions. It is used especially in the field of education.

There’s a Wikipedia article with some technical data on http://en.wikipedia.org/wiki/Flesch%E2%80%93Kincaid_readability_test.

There is information also on http://www.readabilityformulas.com/flesch-grade-level-readability-formula.php. Here you can download a free 62-page PDF booklet on the uses of Readability Formulas.

Send to Kindle