Leaving aside Latin Office Hymns, because I, let's face it, they're all great, even if some were slightly mangled by Urban VIII, and there isn't enough room to list them all.

A good hymn is a marriage of text and tune. The text needs to have a clear rhythm that serves text, and that serves as an aid to memorization, being "artsy" and altering to make a statement breaks the rhythm creates a bad hymn. The text also needs to be direct, to have a statement to make and make it. The tune also needs to have a strong rhythm, it should avoid being clever and "artsy", it cannot be delicate, it has to be able to bear the weight of massed singing by untrained singers, and so should avoid chromatics, melodic tritones, augmented intervals, upward leaps greater than a major sixth, or downward leaps greater than a minor sixth.

At Mass this morning the "big hymn" was "All hail, adorèd Trinity" set to ILLSLEY, number 145 in the New English Hymnal, so I will use this as an example since I'm thinking about it.

The text is a translation by John Chambers (1805-1893) of the 10th century Latin hymn 'Ave colenda Trinitas'. The text is set in clearly defined lines of eight syllables, there are no enjambments or artsy literary devices that obscure the rhythm, the rhymes are clear not clever.

The tune, ILLSLEY, is by John Bishop (c. 1665-1737). The melody is primarily comprised of stepwise movement but with one memorable leap in each phrase. The first two phrases answer each other, both beginning with the same motif of a descending fourth from tonic to dominant and back to the tonic, a typical kind of fanfare figure; the second phrase ending with a half cadence (on the dominant). The third phrase begins stepwise from the dominant to the third, but then leaps up a fourth to the sixth degree of the scale, and lingers there for a bit, before descending stepwise to the second degree of the scale. The final phrase begins primarily stepwise (the third between the second and third notes is ornamented with a passing tone) but has an upward leap on the fourth note from tonic to dominant, then a descending fourth to the supertonic, and ends with stepwise movement and a tenorizans clause at the final cadence. The melody is simple, somewhat intuitive, with some memorable leaps, all of which line up with important words in each verse. Which brings up another point about hymn texts in general: because each verse is sung to the same music, important words should be put in the same place in each verse so that composers have half a chance of writing music that fits the text, without worrying that in verse 7 the word "and" falls on a climactic note and is given too much emphasis.

In both text and tune the sentiment is clear and direct, unencumbered by artistic pretense or ego, neither Chambers nor Bishop are trying to create a magnum opus. This is not a work of the "I" of the authors, but a vehicle for the communal worship of the "we" of the Church.

Some of the newer office hymns are not that good… This is one area where I'm always miffed at what's in the 1912/1949 antiphonal but also not impressed by what came in 1983. I note that for all of the efforts to restore what was perceived as the pristine meter, there are probably more hymns with hypermetrical texts such that you don't really have a choice. There is simply no note in the Solesmes/Vatican editions, whereas the older hymns (mangled or not) leave the note, so you can sing both notes with two vowels, you can sing two notes but one vowel (like "Patre_et almo Spiritu" or in the Veni Creator), or you can sing just one note, depending on the context.

Anyway, the text-to-tune relationship depends somewhat on the language, which is not irrelevant, even on a majority-Anglophone, even) forum, because so many of our best hymns are translations, some of which use the original tune. Would "Holy God, We Praise Thy Name" be the same without GROSSER GOTT? No, I don't think so. Is it always bad to change the tune? No, if it doesn't work at all in English. However, it's easy to find an example of where I think this goes wrong. "Macht Hoch Die Tür" uses a melody of Freylinghausen, which is the second melody used with the tune, and this particular one was originally matched (well, composed for it, I suppose) in the late 17th or early 18th century. But it's beautiful. English speakers usually use TRURO. Frankly, I don't think it's a good match, and you absolutely can sing the melody used in German with the very good English translation. For what it's worth, I think that the Freylinghausen melody does what Salieri does.

On the other hand, some of what I consider the best French hymns have a music-to-text relationship that I find bizarre as a native English speaker. "Chez Nous, Soyez Reine" is a good example. The longer and shorter notes alternate on words that are never the final word in a phrase (and therefore the "accented" word), like in the first verse, where a preposition is accented by virtue of the length of the note, and yet it's an absolute banger.

Sorry, I wish that I was more musically literate such that I could express myself more technically.

Many moons ago I had a hymn blog called Hymnography Unbound.

The sidebar shows some possibly helpful resources, though I haven't checked the links lately, and categories like "good hymns" and "great hymns."

http://hymnographyunbound.blogspot.com/?m=1

The tune GROSSER GOTT was written to fit the German translation ("Grosser Gott, wir loben Dich") of the Te Deum laudamus. "Holy God, we praise Thy Name" (of which, we only sing the first four stanzas nowadays) was not written as a translation of the German text, but rather it was written as an English translation from the Latin of the Te Deum laudamus to fit the tune GROSSER GOTT. See this thread, and my setting of the full English text.

We sang a hymn of impeccable artistry to its tune of exemplary grandeur last evening at Walsingham (solemn Evensong and solemn Te Deum honouring the tenth anniversary of our rector). The hymn is admirable, even cerebral theology with a Latin refrain. The tune breaks all the rules for a 'good' or 'singable' hymn tune - it straddles a tenth and is quite jagged. Around 250 people sang it heartily and with gusto almost raising the roof.

The text, almost cerebral, was 'Firmly I Believe, and Truly'
The tune, of course, was Nashola House


In the light of such a challenging example it is very difficult to pin down what the characteristics of a good hymn/tune are. A wise person will never go 'round announcing that he or she is wise. For him or her to do so we would all know that he or she isn't!
Likewise, we all know (at least all of US) a fine hymn/tune when we hear it, but to put that fineness into words will likely be a fruitless exercise in subjectivity.

To get back to the OP, a few thoughts on what makes up the fabric of a good hymn come to mind:

• Firstly, I echo Salieri that tunes need to be singable, both in terms of melody and rhythm. Overly complicated rhythms (especially, heaven forbid, "hymns" where the rhythm is different for every verse! *cough*SLJ's*cough*) and huge intervals need to be avoided as much as possible. The intervals rule should only be bent in rare circumstances, and only when the melody is easily grasped by the ear, in such a way that people shouldn't have to "think" about how to sing the interval.
• As a subset to the rhythm issue: the natural stresses of the text should align with the natural metric stresses of the chosen tune too. It is very much possible to mis-match texts and tunes based on syllabic stress alone. This can ruin the effect of both and render singing quite uncomfortable, even if neither the text nor the tune present any problems on their own. Imagine: "Alleluia, Sing to Jesus" to O WALY WALY. Nopety, nope, nope!
• A secondary consideration with the melody is that the style of the hymn should match the text too. "Alas and did my Saviour bleed" or "O Sacred Head" should not be married to cheerful hymn tunes just because the meters match. Imagine "O Sacred Head" to ELLACOMBE. Awful. Awful. Awful.
• Hymns should be theologically rich. This can manifest a few different ways: direct quotation or tropes of scripture (the gold standard), settings of prayers (especially ancient prayers from the liturgy or church Fathers), or even tropes of the propers themselves.
• Hymns should not be anthropocentric, but rather theocentric. Enough of the old trope of singing about singing, and yay us for being here!
• If the hymns are prayers, they need to actually petition God for something, and not just talk to/about God generically. If a hymn is focused on the mercy of God, it's probably good to implore that mercy at the end of the hymn rather than just acknowledging that He is merciful.
• Hymns that focus on offering a theological reflection should be soundly rooted in proper Catholic doctrine, and should seek to reinforce it free from any ambiguity. This is most practically applied to hymns which touch Trinitarian doctrine as well as Eucharistic realities. In the case of the latter, terms like "Flesh" or "Body" should be given preference to "bread" (unless its using the ancient term, "Bread of Angels" which is a proper appellation from antiquity) as "bread and wine" without any qualifiers very easily lead to ambiguity, and do nothing to catechize or reinforce the reality of the True Presence. (And Lord have mercy on those hymnographers who have tried to deliberately undermine this truth!)
• All of the above is predicated on the fact that the language used is worthy of the temple, and, whenever possible, elevated. Few things are as dreadful as bad poetry and forced rhymes in pedantic, quotidian language. We are talking formal latria here, not kumbaya around the campfire. IMHO, it is better to stretch the singer with a profound text with a few big words than it is to dumb everything down to 3rd grade levels. God understands, even if a few people with poor vocabularies don't.
• Texts and translations from antiquity should be given preference to novel texts wherever possible. The greater portion of our superlative hymnographical patrimony dates to ancient latin hymns from antiquity, often written by saints/doctors/consecrated religious. Bernard of Clairvaux and Aquinas were better hymn writers than you and I will ever be, and there are many well known translations which are now in the public domain that are more beautiful and poetic than we will ever be able to achieve. (To be clear, this is not to say that there is no place or room for modern hymnographers, of which I consider myself to be but one example.)

Serviam,

Thank you for your succinct presentation of the principles of good hymnody.

Occasionally, schlock slips past the dam you have set up. Sister Suzanne Toolan's I am the bread of Life passes your gold standard (quoting Scripture) and presents no unpredictable or unsingable intervals, and (until Ted Turner got his crayons near it) it was theocentric.

To quote the Bishop of Ely (Henry VAct I, Scene i)

What prevention?

It seems to me that Toolan’s IATBOL fails some of the above tests.

It seems that “the style of the hymn should match the text” fails by a whisker. It's a rousing melody, pretty singable albeit with a wide range, but I cannot think of such rousing melody set to the solemn words of God Incarnate. More fitting for our words, of praise or faith.

It seems that in subsequent verses of IATBOL “the natural stresses of the text should align with the natural metric stresses of the chosen tune” fails rather badly. The hymn is noticeable example of the practice of forcing unmetrical lines into song. Here, “He who comes to Me shall not hunger” and “I am the life” mysteriously are set to the same line of music!

(To be fair, O Come All Ye Faithful does the same thing, "God of God” to the same line as "O come all ye faithful”.)

But perhaps one more test could be added: “the use of repetition should be limited” (with an exception for Gloria and Alleluia).. IATBOL makes us belt out “I will raise him up” 15 times by the end, and it's too much.

I Am the Bread of Life joke verses. I did not create these.

(Purple) But blue is depressing! (Kill the purple guy)

Truly, Beauty does not lie in the eye of the beholder. Often times the brain behind the eye is telling lies to the eye.

I think the rules for choosing liturgical music are contained in what the US bishops have presented as the three judgments: musical, liturgical and pastoral. (cf. Sing to the Lord) Music for Mass should satisfy all three judgments. It's a judgment of art, not an algorithm, so much as Pope Francis has said about liturgical formation in Desiderio Desideravi, the spirit of the liturgy needs to be internalized by all ministers and even by the members of the assembly in order for the liturgy to be celebrated authentically and well.

Far too many people deciding and providing liturgical music at Catholic parishes hardly know what they're doing. OCP and GIA know that, which is why they create menus of songs for people to select from for each Sunday and Solemnity of the year. But those publishers are interested in selling new music, not in promoting what will best serve the Church's liturgy.

The problem with STTL, and other such modren legislation (or quasi-legislation, as STTL is, since it never received recognitio from Rome), is that no one has ever defined what the heck "pastoral" even means, other than when it literally refers to sheep or the English countryside.

I can look at just about anything in a certain missalette published in the North Western U.S., and roundly reject 99% of its contents on musical and liturgical grounds, but then comes the pesky "pastoral" issue. What does it mean? In practical terms, in most people's (clergy and parishioners, and professional liturgists and "liturgists") minds it means "those things which a given community wants, is accustomed to, and likes. And others add that the pastoral considerations trump all others. So, if it is allegedly "pastoral" to sing all of this dreck and twaddle at Mass, however poor it is musically, or however much it deforms the liturgy, because the people know it and it therefore fosters "active participation" (whatever that means at any given time), then we're back to square one.

Or a secondhand bookstore.

I thought the 'my dog's bigge than your dog' was a kennelration commercial!