Where should you start with Lithuanian — words, cases, or ready-made phrases? Short answer: start with phrases and your first words, bring in the cases once you have about 100–150 words, and from then on run all three in parallel. Below is the recipe, stage by stage. Each step shows what it rests on and what it should look like on the "Where your effort goes" radar.
How to read the radar
The radar on the home page splits your points between three sections: Words, Phrases and Grammar. The dark layer is this week, the light one is all time.
Points are not equal: a word is 1 point (3 once learned), a phrase is 1 (3 from stage two), a passed grammar lesson is 5. You touch far more words in a week than you pass lessons, so an even 33/33/33 triangle is not the goal. Words will almost always be the longest axis. Read the percentages in the table under the radar and compare them with the ranges below.
Stage 1. The first two weeks: phrases and first words, no tables
What to do:
- Words: the first lessons of the "Sėkmės!" A1 course in Words — "Koks jūsų vardas?", "Čia mano draugas", "Koks tavo adresas?". That is 64 words.
- Phrases: the "Sėkmės! A1.1 — Phrases 1" programme in Phrases.
- Grammar: not yet.
Why. In a two-year study, Myles, Hooper and Mitchell (1998) followed school beginners learning French. The children first learned phrases whole. Then they gradually "unpacked" them and reused the parts to build new sentences. Learning chunks by rote and working out rules turned out not to be two separate routes but one process: the phrases feed the grammar.
Phrases are also a large part of real speech. Erman and Warren (2000) counted that about half of text is made of prefabricated combinations. Conklin and Schmitt (2008) showed such combinations are read faster than free ones — by native speakers and learners alike.
Norėčiau kavos — "I'd like a coffee". Learn the phrase and the genitive kavos is already inside it. The rule comes later and lands on an example you already own.
On the radar: words 60–70%, phrases 30–40%, grammar 0%. That is fine.
Stage 2. After 100–150 words: singular cases, one at a time
What to do:
- Grammar: the "Lithuanian Cases" programme. Start with the accusative (Galininkas: turiu knygą), then the locative (Vietininkas: gyvenu Vilniuje), then the genitive (Kilmininkas: norėčiau kavos). In parallel, "Verb Conjugation", starting with the present tense.
- Words: keep going with "Sėkmės!" and the topic lists: 10–20 new words a day plus review.
- Phrases: keep going with A1.1.
Why now. A case needs words to decline. Without a vocabulary the table of endings is empty. How many words exactly — no study says. 100–150 is Fluent's guideline: by then you have enough nouns to try every rule on words you already know.
Why teach grammar explicitly. It works. Norris and Ortega's meta-analysis (2000): instruction with direct attention to form has a clear and lasting effect. Spada and Tomita (2010): explicit teaching helps with simple and complex rules alike. Kang, Sok and Han (2019) pooled 35 years of research and confirmed the same.
Why one case at a time. Pienemann (1984), working with children learning German, showed that a structure can be learned under instruction only if the learner is already one step before it. You cannot skip a stage. That is why in Fluent the next lesson opens only once the previous one is passed above 75%.
How often. Grammar needs spaced review too. Bird (2010): distributed practice of syntax beat massed practice on the delayed tests. Suzuki (2017) compared 3.3-day and 7-day gaps for learning morphology. The 3.3-day group was more accurate, and the gap held at 7 and 28 days. In practice: a grammar lesson every 2–4 days, not a weekly marathon.
On the radar: words 55–65%, phrases 20–30%, grammar 10–20%.
Stage 3. Around 500 words: plurals, past tense, mixed review
What to do:
- Grammar: plural cases and the past tense of verbs. After each passed review-level lesson, use "Remind me what I might have forgotten" on the Grammar page — a mixed check across every lesson you have passed.
- Phrases: the "Sėkmės! A1.2 — Phrases 1" programme (377 phrases).
- Words: the whole "Sėkmės!" course — 632 words — then the topic lists.
Why mix now. Pan and colleagues (2019) taught adults to conjugate Spanish verbs in two past tenses. When all training happened in one session, mixing the tenses did not help. When training ran over several sessions, mixed practice gave a substantially better result a week later. The takeaway: learn each form separately first, then mix them in later sessions.
Why not mix words. Brunmair and Richter's meta-analysis (2019) of 59 studies: interleaving helps overall, but for word material, blocked presentation came out ahead. That is why words in Fluent come in topic lists, while mixed review is a grammar feature.
On the radar: words 45–55%, phrases 25–35%, grammar 15–25%.
All the way through: vocabulary never stops
Words stay the longest axis at every stage, and rightly so. Nation (2006) calculated that knowing 98% of the words in a text takes about 8,000–9,000 word families for reading and 6,000–7,000 for speech. Schmitt, Jiang and Grabe (2011) tested 661 people: comprehension rises with the share of known words almost in a straight line, with no "threshold" where everything suddenly clicks. Every new word helps.
Within vocabulary, Fluent picks what to review for you. Testing yourself sticks better than re-reading (Karpicke & Roediger, 2008).
What the radar does not show
Nation (2007) suggests splitting time roughly equally between four strands: meaning-focused input (reading, listening), your own output, deliberate language study and fluency. Everything the radar measures — words, phrases, grammar — is one strand: deliberate study.
So on top of a balanced radar, add what it does not count: read the Articles, listen to Lithuanian, speak. The radar watches the balance inside one part, not the whole language.
When to correct course
- Grammar at 0% for two weeks and you already know over 150 words — open the next case lesson.
- Phrases below 10% — go back to the phrase programme; without ready-made chunks, speech stays slow.
- Words below 40% — grammar will soon run out of material. Add a new list.
- This week looks very different from all time — you can see it in the radar's two layers. One week is fine. Three in a row means your balance has shifted; check whether you meant it to.
The honest limits
No experiment has compared exactly this order with another and measured the difference. Each step rests on research: phrases early, explicit grammar one step at a time, gaps of a few days, mixed review for rules and blocks for words. But the specific numbers — 100–150 words, 500 words, the radar percentages — are Fluent's guidelines, drawn from those studies and from how points are counted. They are not a measured result.
References
- Myles, F., Hooper, J., & Mitchell, R. (1998). Rote or rule? Exploring the role of formulaic language in classroom foreign language learning. Language Learning, 48(3), 323–364. https://doi.org/10.1111/0023-8333.00045
- Erman, B., & Warren, B. (2000). The idiom principle and the open choice principle. Text, 20(1), 29–62. https://doi.org/10.1515/text.1.2000.20.1.29
- Conklin, K., & Schmitt, N. (2008). Formulaic sequences: Are they processed more quickly than nonformulaic language by native and nonnative speakers? Applied Linguistics, 29(1), 72–89. https://doi.org/10.1093/applin/amm022
- Norris, J. M., & Ortega, L. (2000). Effectiveness of L2 instruction: A research synthesis and quantitative meta-analysis. Language Learning, 50(3), 417–528. https://doi.org/10.1111/0023-8333.00136
- Spada, N., & Tomita, Y. (2010). Interactions between type of instruction and type of language feature: A meta-analysis. Language Learning, 60(2), 263–308. https://doi.org/10.1111/j.1467-9922.2010.00562.x
- Kang, E. Y., Sok, S., & Han, Z. (2019). Thirty-five years of ISLA on form-focused instruction: A meta-analysis. Language Teaching Research, 23(4), 428–453. https://doi.org/10.1177/1362168818776671
- Pienemann, M. (1984). Psychological constraints on the teachability of languages. Studies in Second Language Acquisition, 6(2), 186–214. https://doi.org/10.1017/S0272263100005015
- Bird, S. (2010). Effects of distributed practice on the acquisition of second language English syntax. Applied Psycholinguistics, 31(4), 635–650. https://doi.org/10.1017/S0142716410000172
- Suzuki, Y. (2017). The optimal distribution of practice for the acquisition of L2 morphology: A conceptual replication and extension. Language Learning, 67(3), 512–545. https://doi.org/10.1111/lang.12236
- Pan, S. C., Tajran, J., Lovelett, J., Osuna, J., & Rickard, T. C. (2019). Does interleaved practice enhance foreign language learning? The effects of training schedule on Spanish verb conjugation skills. Journal of Educational Psychology, 111(7), 1172–1188. https://doi.org/10.1037/edu0000336
- Brunmair, M., & Richter, T. (2019). Similarity matters: A meta-analysis of interleaved learning and its moderators. Psychological Bulletin, 145(11), 1029–1052. https://doi.org/10.1037/bul0000209
- Nation, I. S. P. (2006). How large a vocabulary is needed for reading and listening? The Canadian Modern Language Review, 63(1), 59–82. https://doi.org/10.3138/cmlr.63.1.59
- Schmitt, N., Jiang, X., & Grabe, W. (2011). The percentage of words known in a text and reading comprehension. The Modern Language Journal, 95(1), 26–43. https://doi.org/10.1111/j.1540-4781.2011.01146.x
- Karpicke, J. D., & Roediger, H. L. (2008). The critical importance of retrieval for learning. Science, 319(5865), 966–968. https://doi.org/10.1126/science.1152408
- Nation, P. (2007). The four strands. Innovation in Language Learning and Teaching, 1(1), 2–13. https://doi.org/10.2167/illt039.0