Duplicated rows when aggregating data in dplyr() - r

Question

I'm trying to create a set of cross-linguistic data by joining three datasets together in dplyr(). Two of the datasets are 'dictionaries' of sorts - they are word lists that I want to attach to speakers. There are 15 speakers and so a number of repetitions throughout the data, while each word only appears once in each of the dictionaries.
When I join two using left_join(), I get replicated cells. I know I can remove the duplicated cells, but I sense that there must be something simple that I'm doing wrong to create this issue.
Example data is as follows:
French <- c("un", "deux", "trois", "chien")
English <- c("one", "two", "three", "dog")
type <- c("number", "number", "number", "animal")
speaker <- c(1, 1, 1, 4)
df.fr = data.frame(speaker, French)
df.en = data.frame(speaker, English)
df.type = data.frame(English, type)
I want to create a new dataset, new.df, by joining df.en and df.fr by speaker, and then joining that to df.type by English.
Preferably I would use dplyr() to do this. When I do the following, I get duplicated rows:
new.data <- df.fr %>% left_join(df.en)
which generates
speaker French English
1 1 un one
2 1 un two
3 1 un three
4 1 deux one
5 1 deux two
6 1 deux three
7 1 trois one
8 1 trois two
9 1 trois three
10 4 chien dog
When really I just want it to join 'un' to 'one', 'deux' to 'two', etc:
speaker French English type
1 1 un one number
2 1 deux two number
3 1 trois three number
4 4 chien dog animal

Answer 1

Aside from cbinding the three datasets, you can create a unique id for each speaker for both df.fr and df.en and join on speaker + id:
library(dplyr)
df.fr %>%
group_by(speaker) %>%
mutate(id = 1:n()) %>%
left_join(df.en %>% group_by(speaker) %>% mutate(id = 1:n()),
by = c("speaker", "id")) %>%
left_join(df.type) %>%
select(-id)
If you have more than two language datasets, you can also write a more general solution using map and reduce from purrr:
library(purrr)
list(df.fr, df.en) %>%
map(~ group_by(., speaker) %>% mutate(id = 1:n())) %>%
reduce(left_join, by = c("speaker", "id")) %>%
left_join(df.type) %>%
select(-id)
Result:
# A tibble: 4 x 4
# Groups: speaker [2]
speaker French English type
<dbl> <fctr> <fctr> <fctr>
1 1 un one number
2 1 deux two number
3 1 trois three number
4 4 chien dog animal