analysis
details
The Grimm Fairy Tales can be analyzed in a multitude of ways, but for our purposes with this project we have focused on a comparison between the German and English translations, with a deeper analysis on female characters. We focus on how the different types of women characters act and how they receive action, as well as the dialogue that they speak.
more details
Below are graphs that have been derived from our data collection that has been curated through our XML markup of the tales. Since this is a group project, the markup tasks were split between project members. This means that each person marked their tales in an original way, and because of this there may be inconsistencies within our data.
These graphs came from research through XPath in our marked up tales. The tales above have a comparison between a mother and stepmother figure, with some exceptions. Through this data collection and the creation of these graphs it is clear that much of the dialogue spoken by the women above is not positive, leaving conversations to mostly be negative or neutral towards other characters. The positive dialogue does exist from the mother in Cinderella, but her dialogue is minimal because she passes away early in the tale. This means that her positivity did not directly reign on Cinderella, although her mother was always looked at in a positive light. The positive dialogue from the Witch in Hansel and Gretel and the Second Queen in Little Snow White occurs mainly as manipulation and not true positivity.
This graph represents main women characters within the tale minus the widow. This comparison is important because it shows women figures that are children as well as an adult figure. The results are interesting because of the way that our dialogue has been marked up. The Beautiful Girl was looked at as a positive character, at least in the English verison, but all of her dialogue is displayed negatively. This is because, while she was a good person, many of her dialogue is, indeed, negative. The Lazy Daughter spoke negatively and was a negative person as well. Frau Holle displayed kindness and positivity towards the Beautiful Daughter because of her willingness to help. She was only negative towards the Lazy Daughter because she was, indeed, lazy.
This graph was only created with the English versions because the majority of our efforts were put towards the English mark up, which resulted in less acts being marked up in the German versions. While there isn’t a comparison from English to German, there is still a lot of data. It is clear that all tales have all types of effects and some have more of one than another, but it is difficult to draw specific conclusions about these results. It may be possible that negative women figures in some of these tales are stronger than others.
topic modeling
English Topics
0 0.25 hansel gretel bread eat father witch pebbles moon evening duckling oven fat stall crumbs deeper full midday pockets awoke wind
1 0.25 mother give sat ate head sister daughter red put killed room crying table stay brother neck gathered green picked stood
2 0.25 beautiful time day thought child looked long asked called fell heart eyes evening found finally dear eat ran frightened happened
3 0.25 home house good white answered back inside leave brought entire knocked sleep died wild climbed put bed true jumped poor
4 0.25 man hand began wanted left threw stood house singing longer cried gave stopped door sun finished song husband high made
5 0.25 heard window standing die world years joy great loved birth peace lived bring life top desire son fear answered began
6 0.25 twelve brothers benjamin tree eleven coffins redeem dear forehead star youngest speaking wood fire happily ravens clothing bewitched flag shavings
7 0.25 woods children night find large made run ground spoke beds small carried harm shook comforted remained feathers rid great stake
8 0.25 red cap grandmother wolf wine big open grandmother's cake flowers don't trough wolf's path loudly stepped i'm live baked taking
9 0.25 father beautiful wife flew cried happy golden sitting don't shoes blood passed death hear pot evil snow finished lit birds
10 0.25 snow-white mirror dwarfs queen woman fairest poisoned wall land mountains coffin snow-white's huntsman dead comb snow apple sale fairer black
11 0.25 bird marlene tree juniper tweet beneath sing apple scarf bones laid tied chain fell silken sang feel boy clickety-clack month
12 0.25 rapunzel sorceress hair tower prince climbed garden wife woman voice climb miserably sorceress's wall wandered beloved wilderness happening gothel frau
13 0.25 cinderella pick prince shoe dance bride hazel festival dress clothes grave goo rook horse danced sky sisters waited coop turned
14 0.25 king queen shirts brothers speak swans castle hunting laugh word lifted show yarn ball single mouth touched accused swan-skins boys
15 0.25 roland magic sweetheart bed witch flower long married stepdaughter middle faithful wise shepherd steps wand i'm hedge duck happened shouted
16 0.25 door morning lying piece asleep water wood i'll afterward apron back poor opened early pulled stuck arrived jumped half kill
17 0.25 girl wicked told blood shouted walked side hour god wedding thinking face work forced crawled straight bedroom beauty days scattered
18 0.25 tree pigeons stepmother ashes bird foot kitchen pigeon silver cut gold beneath gave began turtledoves throw lentils dressed man wept
19 0.25 woman fire roof front cry shining set stone answer ears father's flames open corner mind dropped walking good-bye seated shoemaker
German Topics
0 0.25 gretel nsel brod weg vater cklein essen sagte fett freute mond wasser uschen sagen sprach uslein paar all seh tasche
1 0.25 wollte ward stand tag konnte ber haus nen sprach kind ten endlich nacht fing bleiben ger lieb darnach tochter hen
2 0.25 gro kamen geh gewesen gekommen sagten gedachte fiel dern arbeit prinzessin hrte einander herz thun ging ter backen holte gemahlin
3 0.25 prinz che braut treppe umlein kleider rechte tanzen guten sche ttel herab puttel lteste ren gesicht unten mitternacht schnitt chenput
4 0.25 sechs nig weg brachte prinzen stumm schw warf hrt zeigte nne schwiegermutter jahren knauel verwandelt linke wickelte daf zweiten lebte
5 0.25 blut sen bat gerade wenns neid fertig giftigen haushalt nheit setzten stellte geworden hervor bettlein gaben aussah hrten fahne verl
6 0.25 rapunzel fee mann frau haare thurm herunter haar rapunzeln tages erschrack leiter thr nigssohn mmerlich stenei abge ritsch rapunzels worte
7 0.25 dchen nig leben welt schwesterchen ohne baum nigin jagd verloren hinauf hohen wort fielen hemder erl hle hemdlein luft buben
8 0.25 lie ren rte riefen lassen angst fest stiefmutter daheim freuden federn nieder solle finden tausend rothen menschen bekam nnen hne
9 0.25 sagte haus rief wei chen recht fort ganz heim ner roth kopf tausendmal gen schlief guckte trat halten macht wolle
10 0.25 frau gut sch holle ach ging flei thor bett aepfel alte raus zieh wiese wasser lebtag bedeckt ngst faule ckte
11 0.25 wald seyn sehen gingen kochen rothe gefressen liebe hren walde kleine zufrieden standen schwer dchens glaubte gute bsch verwunderte guckst
12 0.25 feuer kinder morgen alte hexe mitten tte abend setzte lein morgens holz tod frau fortgegangen verbrannt genug bettlein kin gedeckt
13 0.25 machte wollen abends augen tage dach fenster hand voll sonne gott kindern los hinab garten hals guter kieselsteine fen gewaltig
14 0.25 mutter sagte zwei still stand guck herzen that rein acht geschwind ach gehen ans schornstein stra welt satt letzten tel
15 0.25 sneewittchen nigin sieben frau nste spieglein spiegel land sarg schn wand seyd kamm sneewitt zwerge liegen schnee gegessen alte essen
16 0.25 bett sprang roland blume lief liebsten lang sten weit gekocht gethan dchen dar eigen vorne suchte traurig gingen stief weise
17 0.25 gro mutter ppchen rothk wolf bring wein gedacht kuchen blumen weg trog steine leib schwach schnarchen fressen nge haube klinke
18 0.25 sch ging sah sey hinaus todt lange men zeit fragte komm her ben dar stube legte hlte drei sehen lebendig
19 0.25 chenputtel pantoffel tauben schwestern wagen pick pfchen linsen schuck rucke hrte lesen heerd grab schwe ball ckte toffel pan goldenen
topics analysis
When utilizing topic modeling for this project, the team envisioned that the results would point to what the tales are about, getting at themes of positivty or negativity, values and morals, and other things that may not be immediately visible to analysis. With this hope in mind, we ran topic modeling, and instead of receiving data that touched on broad themes, the exact opposite happened. The topics that were produced were strikingly similar to individual tales, so much so that they don't speak to the corpus we analyzed as a whole. For example, topic 0 of the English topics is a condensed version of Hansel and Gretel, and topic 3 of the German topics is a condensed version of Aschenputtel(Cinderella). Because of this, the topics that were created, but do not seem to be productive in furthering analysis on our research question. However, it does provide some insights as to what the tales are invested in, and how they may differ.
What was made clear through the topic modeling is that the tales are heavily invested in the characters themselves, and it is the named characters and individuals found within them that make up a bulk of the differences one can find. This is striking, though understandable, as the tales themselves often speak to similar themes and work in similar ways. Part and parcel of these tales is that young people are put into precarious situations, often due to some break in kinship (such as the evil mother/stepmother figure) and come out of the situations matured and with some degree of autonomy. The difference between the tales depends on the individual characters and the particularities of their actions and behaviors. Therefore, seeing the topics hone in on those individual characters makes clear sense.
However, if this project was not created and executed within the timeframe of a semester, there are avenues of change that could make topic modeling possibly more productive. For example, it could be productive to standardize all proper nouns in the tales, stripping the named individuals of that particular difference so that topic modeling software doesn't find that "Hansel" or "Gretel" is useful data. Another method could be to work with kinship terms and characters that function as kin. For example, in the tales in our selected corpus, individuals named as stepmother, queen, and sorceress all function as the archetypal evil stepmother. By collapsing all of these characters into one term, topic modeling could more easily dive deeper into the texts and provide models that do not give surface-level analyses of difference.
While the team hoped for more with topic modeling, we are happy to see work done with this technology and the potential it provides. We are also glad use the data we do have to draw some conclusions, mainly that the tales do not differ strongly; they seem to follow similar themes and narratives due to the topics' investment in named characters instead of other difference. Because of this insight, the team has chosen to include our findings.